Immunization in the Acute Care Setting: Pneumonia Prevention in Utah Elderly

 

by

Juli A. Antonow, M.D., M.H.A. and Michael P. Silver, M.P.H.

 

Abstract

In 1996, pneumonia and influenza together ranked fifth as a leading cause of death in Utah.  Streptococcus pneumoniae and influenza A are the major causes of preventable pneumonia.  Rising incidence of antibiotic-resistant Streptococcus pneumoniae is also a threat; prevention of pneumonia is becoming more important as it becomes more difficult to treat.  Vaccines are available for both these infectious etiologies, but though efficacious and cost-effective, are underutilized.  Both vaccines are recommended for all persons age 65 or greater; data shows that those hospitalized are an especially high-risk group for rehospitalization for pneumonia.  Hospitalizations in this age group should be considered opportunities to vaccinate a high-risk group to prevent subsequent pneumonia hospitalizations.  Effective strategies for improving vaccine use in inpatient settings are system-oriented interventions such as standing orders or policies.  Hospitals should include immunization of elderly inpatients using proven interventions to ensure vaccine delivery.

 

 

 

Problem Statement--Pneumonia in Utah

            In 1995 and 1996, pneumonia and influenza together ranked sixth nationally among the leading causes of death; more than 90% of these deaths were in elderly adults (>65 years of age) (CDC, October 1997). In Utah, the ranking was fifth in 1996 with a rate of 23.7 deaths per 100,000 due to pneumonia and influenza compared to a national rate of 31.1 (Utah Department of Health, 1996). There were 475 deaths in Utah in 1996 due to pneumonia and influenza; 423 (89%) were in persons aged 65 years and above (Utah Department of Health, 1996). That year, in a population of approximately 170,000 Utah Medicare beneficiaries aged 65 and over, 5.7% (n=2381, Table 1) of all Medicare admissions to Utah acute care facilities were for a pneumonia-related principal diagnosis. This represents, with allowed charges paid to hospitals of $5,400 per case, $12,800,000 in costs to Medicare (not including reimbursement to physicians by Medicare). Utah Medicare beneficiaries admitted to acute care facilities for pneumonia in 1996 experienced 7.2% inpatient mortality, with mortality rising to 14.3% at 30 days post-discharge (Medicare Part A billing data analysis by HealthInsight, unpublished).

Text Box: Table 1. Medicare Part A-Derived Statistics on Utah Medicare Beneficiaries* 1993 1994 1995 1996 Utah Medicare beneficiaries age 65+ (with Medicare Part A coverage 7/1/9x) 158,848 162,669 166,209 169,385 Number of hospitalizations (Utah facilities, Medicare beneficiaries age 65+, any state of residence, any diagnosis) 38,988 38,777 40,943 42,128 Number of Utah Medicare beneficiaries age 65+ hospitalized 25,586 25,554 26,650 27,584 Number of pneumonia admissions (Utah hospitals, Utah Medicare beneficiaries age 65+) 2,403 2,159 2,529 2,381 Percent pneumonia admissions discharged alive followed within one year (and after 30 days) by another pneumonia admission (age 65+, overall=10.2%) 9.9% 9.3% 11.1% 10.3% * Each admission with a live discharge counts as a separate event towards the total readmission percent, even if a patient has more than one readmission for a pneumonia-related condition.
In Utah, pneumonia is also an important rural health issue. Ambulatory care sensitive (ACS) hospitalizations are felt to be avoidable with optimal preventive care and early disease management in the outpatient setting. ACS hospitalization rates for respiratory conditions (including pneumonia) for Medicare beneficiaries in rural Utah counties indicate statistically significantly higher hospitalization rates compared to urban counties for a 5 year period (1990-94) (Silver, Babitz and Magill, 1997). The authors speculate that barriers in access to preventive care in rural Utah may contribute to excess hospitalizations for pneumonia.

 


Preventable Pneumonia Etiologies

Streptococcus pneumoniae (pneumococcus), a bacteria, accounts for an estimated 500,000 cases of pneumonia yearly in the United States. Adults > 65 years are at increased risk; adults with compromised immunity and chronic medical conditions are at even greater risk (CDC, April 1997). Influenza A viruses cause yearly seasonal epidemics. The elderly and those with underlying health conditions are at two-to-fivefold greater risk from influenza complications such as pneumonia requiring hospitalization. Estimated influenza-associated deaths ranged from 20,000-40,000 during each of 10 influenza epidemics from 1972-1991. Influenza-associated deaths result from pneumonia or by exacerbation of underlying cardiopulmonary and chronic conditions, more common in the elderly (CDC, May 1996). Vaccines are available which are effective in the prevention of these two etiologies for pneumonia in the elderly.

 

Rising Incidence of Antibiotic-Resistant Pneumonia-causing Bacteria

Data on the incidence of infections due to antibiotic-resistant Streptococcus pneumoniae are not available in Utah because only outbreaks are reportable according to the current Utah Communicable Disease Rule. In 1995, the Council for State and Territorial Epidemiologists and the Centers for Disease Control and Prevention (CDC) recommended that states require reporting of antibiotic-resistant Streptococcus pneumoniae. Currently 19 states comply, and a revision of the Utah Communicable Disease Rule has been proposed that would make infections with antibiotic-resistant Streptococcus pneumoniae reportable.

The Surveillance NetworkJ Database--USA, a national database (unpublished, MRL Pharmaceutical Services, 6/1/98) of results from selected laboratories, reports decreasing incidence of penicillin-susceptible strains of Streptococcus pneumoniae from 87.3% in 1992 to 56.9% in 1997. The Pacific Mountain region, which includes Utah, reports only 50.4% penicillin-susceptible strains (213 isolates from all sources) for the first 6 months of 1997. Primary Children's Medical Center, a tertiary care pediatric hospital in Salt Lake City, Utah, reports 76% penicillin-susceptible strains in 1993, and only 59% in 1997 (unpublished, Primary Children's Medical Center microbiology laboratory).

Traditional defenses against drug-resistance have been development of new antibiotics, targeted surveillance, and isolation. However, with Streptococcus pneumoniae, the resistance is arising in the ambulatory setting and subsequently becoming a problem in acute care settings (McGowan and Tenover, 1997), so isolation as a strategy is less useful. Recent strategies have been targeted at the outpatient setting for more rational prescribing of antibiotics for the treatment of childhood infections, often viral. With the rising incidence of drug-resistant bacteria, prevention of infection is more important than ever.

The CDC (Drug-Resistant Streptococcus pneumoniae Working Group) recommends a strategy of targeted vaccination programs to regions with high levels of antibiotic-resistant organisms (facilitated by improved detection using the latest penicillin-resistance screening techniques by laboratories and mandatory reporting of infections with drug resistant bacteria). Vaccination should also be targeted to persons at high risk for infection (possibly including children > 2 years of age in child care centers). The Working Group also promotes judicious use of antimicrobial drugs (avoiding broad spectrum agents, using proper dose and duration, and following established recommtableendations for chemoprophylaxis) (CDC, February 1996).

 

Vaccines--Effective, and Underutilized

Studies have shown effectiveness of the pneumococcal pneumonia vaccine in immunocompetent persons >65 years to be 75%. The vaccine is extremely safe, with minimal local side effects and rare systemic reactions  (CDC, April 1997). Medicare Part B began reimbursing for the pneumococcal vaccine in 1981. The vaccine is indicated once for all persons after age 65, at least 5 years after a previous administration if given before 65 years of age. Revaccination is currently not recommended except for those at highest risk for infection (patients with asplenia, for example) and those who are most likely to have a rapid decline in protective antibody levels because of underlying conditions (such as nephrotic syndrome, renal failure or renal transplant patients) (CDC, April 1997). Studies are ongoing which suggest efficacy of the vaccine declines over 5-10 years (Shapiro, et al., 1991).

Influenza vaccine is recommended yearly by the Advisory Committee on Immunization Practices for all persons >65 years of age (contraindications include only a history of anaphylaxis to eggs or the vaccine).  Significant side-effects are rare. When the vaccine is matched to the epidemic viral strains that year, and when vaccination rates of high risk persons are high, vaccination is currently the most effective measure for reducing the morbidity and mortality of influenza (CDC, May 1996). Influenza vaccination became a covered benefit under Medicare Part B in 1993. Fedson, et al. (1993) reported that influenza vaccination in a case-control study of all non-institutionalized adults aged > 45 years (cases=5,036, each matched with 3 controls, 70-75% were > 65 years) in Manitoba, Canada in 1982-83, and 1985-86, prevented 32-39% of hospital admissions with pneumonia and influenza. They estimated 43-65% effectiveness in preventing deaths from influenza-associated conditions.

Multiple studies have also demonstrated that these vaccinations are cost-effective. Fieback and Beckett (1994) report a cost-effectiveness analysis for influenza vaccine which showed a cost of $145 per year of life gained. Nichol et al. (1994) performed a serial cohort study of 75,000 persons > 65 years of age enrolled in a health maintenance organization in Minnesota in 1990-1993. Influenza immunization rates ranged from 45-58%; vaccination was associated with decreased hospitalization rates (by 48-57%, p=0.002) for pneumonia and influenza, and statistically significant reductions in mortality (by 39-54%, p=0.001). They calculated direct yearly savings of $117 per person vaccinated, a cumulative savings of $5 million. Sisk et al. (1997) performed a base case analysis from three geographic areas in the U.S. By extrapolation, they report that if the 23 million unvaccinated elderly persons in 1993 nationwide had been vaccinated against invasive pneumococcal disease, 78,000 years of healthy life would have been gained and $194 million in health care costs would have been saved.

It is well documented that these vaccines are underutilized. A Healthy People 2000 goal is to increase pneumococcal and influenza vaccination levels to at least 60% for non-institutionalized high-risk persons including those aged >65 years (PHS, 1991). Nationally, progress has been made toward this goal; immunization levels have increased among adults > 65 years (52% influenza, 28% pneumococcal pneumonia) (PHS, 1995). Only 42% of persons in Utah > 65 years of age report ever receiving a pneumonia shot; 70% report an influenza shot during the previous 12 months (CDC [Behavioral Risk Factor Surveillance Survey coordinators], October 1997).

The Advisory Committee on Immunization Practices recommends that the pneumococcal pneumonia vaccine be given during the inpatient stay to eligible patients, but this occurred less than 1% of the time in 12 western states in 1994 (random sample of 5048 hospitalizations for Medicare patients > 65 years with principal diagnosis of pneumonia, 6.2% were in Utah). In addition, medical record review for this study revealed that previous vaccinations were documented in hospital admission histories only 4.7% of the time (CDC [Houck, Lowery, Prela], October 1997).

Reasons for underuse of these vaccines in any setting range from misunderstanding of their effectiveness (although studies show that doubts about effectiveness are common even among vaccinees), contraindications, and side-effects by patients and physicians (physicians see a barrier to immunizing inpatients because vaccine effects such as fever could extend the length of stay). Patients= fear of side-effects was the biggest barrier cited in multiple studies (Fieback and Beckett, 1994). A physician recommendation to be vaccinated has been shown to be an important factor in being vaccinated, but provision of preventive health interventions by physicians has been shown to be variable and suboptimal (Cohen, 1994). Having a scheduled appointment with a physician during the fall season has been shown to be a positive determinant of influenza vaccinations (Fieback and Beckett, 1994). Lack of awareness of recommendations, vaccine delivery and availability problems, and inadequate financing mechanisms to support adult immunization delivery are addidtional barriers (Fedson, 1994). 

 

Medicare-aged Inpatients are at High Risk for Subsequent Pneumonia

Part A Medicare data has unique patient identifiers to enable patient-specific analysis over time. While immunization for pneumococcal pneumonia and influenza is recommended for all Medicare-aged persons, persons > 65 years of age who have been hospitalized are an especially high risk group for rehospitalization for pneumonia. Medicare inpatients (any reason for admission) are roughly three times more likely to be rehospitalized for pneumonia in the following year than Medicare beneficiaries not hospitalized. Utah Medicare patients initially hospitalized for pneumonia (1993-96), experienced a seven times greater risk of readmission for pneumonia within one year compared to Medicare beneficiaries who were not hospitalized (see Table 1).

 

Inpatient Admission--Missed Opportunity for Vaccination

In Manitoba, analysis of claims data from 1982-83 demonstrated that among those >65 years of age hospitalized with influenza-associated diagnoses, 39-46% had been hospitalized during the previous vaccination season; 62-67% of those who died had been hospitalized within the previous vaccination season (Fedson, Wajda, Nicol, and Roos, 1992). A hospitalization should thus be considered an opportunity for vaccination and pneumonia prevention.

Similarly, of Utah Medicare patients hospitalized for pneumonia during the influenza season (January through April, 1994-97), over one-third had been hospitalized during the previous year, almost half of them during the influenza vaccine availability period. These represent missed opportunities to prevent two of the most important etiologies for elderly pneumonia hospitalizations (see Table 2).

 

Text Box: Table 2. Inpatient Admission as a Missed Opportunity for Vaccination* 1994 1995 1996 Percent pneumonia admissions occurring during January-April preceded by any admission during previous year (overall=32.9%) 32.7% 31.8% 34.4% Percent pneumonia admissions occurring during January-April preceded by any admission during October-December of previous year (overall=13.7%) 12.9% 12.5% 15.6% *Utah facilities only, Utah beneficiaries age 65+.
Strategies for Improving Vaccine Use

Many strategies have been tested for effectiveness in inpatient and outpatient settings to improve the delivery of influenza and pneumococcal vaccine to elderly adults. Gyorkos et al. (1994) reviewed 377 published studies which reported effectiveness in vaccine delivery. Twenty-four publications on influenza vaccine and 11 on pneumococcal vaccine met the selection criteria (use of control groups, studies done on humans in developed countries). The target population was mostly >65 years of age; study designs, settings, and interventions varied, but the largest intervention effects were seen in the inpatient setting with system-oriented interventions such as standing orders or policies and procedures to ensure immunization.

A hospital-based, system-oriented intervention consisted of an infection control nurse screening patients on the day of discharge for candidacy for pneumococcal vaccine, providing information about the vaccine, and administering the vaccine. Compared to a control group on another inpatient unit who was not offered the vaccine,  a 78% increase in pneumococcal vaccine delivery was achieved (p=0.001) (Klein and Adachi, 1986). In another hospital-based intervention with standing orders for influenza vaccine (which consisted of orders for the patient nurse to assess the need for the vaccine, review the indications, and administer the vaccine), 95.2% of patients were offered vaccination compared to 22% of patients when physicians were exposed only to chart reminders, and less than 12% of patients when education strategies were offered to physicians (Crouse, Nichol, Person, Grimm, 1994). Fieback and Beckett (1994) also found in their review that interventions targeted at inpatients with standing orders and patient education and follow-up achieved vaccination results of 75-95%.

 

Use of Medicare Data to Measure Improvement Outcomes

If standing orders are put in place, Utah Part B Medicare claims data can be used to determine whether they result in increased vaccination delivery. These data can be provided to individual hospitals to indicate the success of their interventions. Part A data can be used to determine whether the ultimate desired outcomes of decreased pneumonia hospitalizations and pneumonia-related mortality occur, for a hospital population, the Utah Medicare population, or for a specific patient.

 

Conclusion

The potential for every health care encounter to include disease prevention interventions has so far not been realized.  Acute care facilities should expand traditional prevention activities, like health fair screenings, to include the acute hospitalization episode. A hospitalization for Medicare beneficiaries is an opportunity for vaccination for the preventable causes of pneumonia in a high risk population. System-oriented interventions, such as standing orders, have the best effectiveness in improving vaccine delivery.

 

References

CDC. (February 16, 1996). Defining the Public Health Impact of Drug-Resistant Streptococcus Pneumoniae: Report of a Working Group. Morbidity and Mortality Weekly Report, 45(RR-1), 1-20.

 

CDC. (May 3, 1996). Prevention and Control of Influenza: Recommendations of the Advisory Committee on Immunization Practices. Morbidity and Mortality Weekly Report,  45(RR-5), 1-25.

 

CDC. (April 4, 1997). Prevention of Pneumococcal Disease: Recommendations of the Advisory Committee on Immunization Practices. Morbidity and Mortality Weekly Report, 46(RR-8), 1-25.

 

CDC. (October 3, 1997). Missed Opportunities for Pneumococcal and Influenza Vaccination of Medicare Pneumonia Inpatients--12 Western States, 1995. Morbidity and Mortality Weekly Report, 46(39), 919-923.

 

CDC. (October 3, 1997). Pneumococcal and Influenza Vaccination Levels among Adults Aged >65 Years--United States, 1995. Morbidity and Mortality Weekly Report, 46(39), 913-919.

 

Cohen, S.J., Halvorson, H.W., & Gosselink, C.A. (1994). Changing Physician Behavior to Improve Disease Prevention. Preventive Medicine 23, 284-291.

 

Crouse, B.J., Nichol, K., Peterson, D.C., & Grimm, M.B. (1994). Hospital-Based Strategies for Improving Influenza Vaccination Rates. Journal of Family Practice, 38(3), 258-261.

 

Fedson, D.S. (1994). Adult Immunization--Summary of the National Vaccine Advisory Committee Report. The Journal of the American Medical Association, 272(14), 1133-1137.

 

Fedson, D.S., Wajda, A., Nicol, J.P., Hammond, G.W., Kaiser, D.L., & Roos, L.L. (1993). Clinical Effectiveness of Influenza Vaccination in Manitoba. The Journal of the American Medical Association, 270(16), 1956-1961.

 

Fedson, D.S., Wajda, A., Nicol, J.P., & Roos, L.L. (1992). Disparity Between Influenza Vaccination Rates and Risks for Influenza-Associated Hospital Discharge and Death in Manitoba in 1982-1983. Annals of Internal Medicine, 116, 550-555.

 

Fieback, N., & Beckett, W. (1994). Prevention of Respiratory Infections in Adults--Influenza and Pneumococcal Vaccines. Archives of Internal Medicine, 154, 2545-2557.

 

Gyorkos, T.W., Tannenbaum, T.N., Abrahamowicz, M., Bedard, L., Carsley, J., Franco, E.D., Delage, G., Miller, M.A., Lamping, D.L., & Grover, S.A. (1994). Evaluation of the Effectiveness of Immunization Delivery Methods. Canadian Journal of Public Health, 85(Supplement 1), S14-S29. 

 

Klein, R.S., & Adachi, N. (1986). An Effective Hospital-Based Pneumococcal Immunization Program. Archives of Internal Medicine, 146, 327-329.

 

McGowan, J.E., & Tenover, F.C. (1997). Control of Antimicrobial Resistance in the Health Care System. Infectious Disease Clinics of North America, 11(2), 297-311.

 

Nichol, K.L., Margolis, K.L., Wuorenma, J., & Von Sternberg, T. (1994). The Efficacy and Cost Effectiveness of Vaccination against Influenza among Elderly Persons Lining in the Community. New England Journal of Medicine 331, 778-84.

 

Public Health Service. (1991). Healthy People 2000: National Health Promotion and Disease Prevention Objectives (DHHS publication No. (PHS)91-50212). Washington, DC: U.S. Department of Health and Human Services, Public Health Service.

 

Public Health Service. (1995). Healthy People 2000: Midcourse Review and 1995 Revisions. Washington, DC: U.S. Department of Health and Human Services, Public Health Service.

 

Shapiro, E.D., Berg, A.T., Austrian, R., Schroeder, D., Parcells, V., Margolis, A., Adair, R.K., & Clemens, J.D. (1991). The Protective Efficacy of Polyvalent Pneumococcal Polysaccharide Baccine. New England Journal of Medicine, 325, 1453-1460.

 

Silver, M.P., Babitz, M.E., & Magill, M.K. (1997). Ambulatory Care Sensitive Hospitalization Rates in the Aged Medicare Population in Utah, 1990 to 1994: a Rural-Urban Comparison. The Journal of Rural Health, 13(4), 285-294.

 

Sisk, J.E., Moskowitz, A.J., Whang, W., Lin, J.D., Fedson, D.S., McBean, A.M., Plouffe, J.F., Cetron, M.S., & Butler, J.C. (1997). Cost Effectiveness of Vaccination against Pneumococcal Bacteremia among Elderly People. The Journal of the American Medical Association, 278, 1333-1339.

 

Utah Department of Health Office of Public Health Data. (1996). Utahs Vital Statistics: Births and Deaths, 1996. Salt Lake City: Bureau of Vital Records, Utah Department of Health.

 

 

About the Authors

Juli Antonow, M.D., M.H.A. is a faculty member in the Division of General Pediatrics, Department of Pediatrics, at the University of Utah. Her roles include positions as Medical Director of the Infant Unit at Primary Children's Medical Center and Utah Medical Director and Principal Clinical Coordinator at HealthInsight.

Michael P. Silver, M.P.H. received his B.S. in Mathematics and M.P.H. from the University of Utah. He is currently Senior Health Care Analyst at HealthInsight, a non-profit health care quality improvement organization in Utah and Nevada.

 

 


[1] Death and birth rates are calculated based on all identified events rather than a sample of such events.  Nevertheless, rates are calculated based on an arbitrary unit of time and will exhibit variability from interval to interval.  That is, the calculated rate can be thought of as a Asample@ of the underlying true rate.

[2] BRFSS is a random-digit-dialed telephone survey of the adult noninstitutionalized population, aged 18 or older.  It is a multistage cluster sample.

 

2 It is a complex survey sample designed to be representative of all Utahns (Utah Bureau of Surveillance and Analysis, 1997).


[i]. Public Employees Health Plan, which administers health insurance for many Utah public employees. Benefits are less comprehensive than Medicaid benefits.

[ii]. HEDIS is the acronym for the Health Plan Employer Data and Information Set.

[iii]. Data available from the authors upon request.