Published online February 1, 2006
PEDIATRICS Vol. 117 No. 2 February 2006, pp. 176-180 (doi:10.1542/peds.2005-0752)
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow P3Rs: Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when P3Rs are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow E-mail this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My File Cabinet
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via CrossRef
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Gorton, C. P.
Right arrow Articles by Jones, J. L.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Gorton, C. P.
Right arrow Articles by Jones, J. L.
Related Collections
Right arrow Infectious Disease & Immunity

Wide Geographic Variation Between Pennsylvania Counties in the Population Rates of Hospital Admissions for Pneumonia Among Children With and Without Comorbid Chronic Conditions

Christopher P. Gorton, MD, MHSAa,b,c and Jayne L. Jones, MPHa

a Pennsylvania Health Care Cost Containment Council, Harrisburg, Pennsylvania
b Department of Health Policy, Jefferson Medical College, Philadelphia, Pennsylvania
c Department of Health Evaluation Sciences, Pennsylvania State University, College of Medicine, Hershey, Pennsylvania


   ABSTRACT
TOP
 ABSTRACT
METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 
OBJECTIVES. To compare county rates of hospital admissions for pediatric pneumonia and to assess the contribution of comorbid chronic conditions to county and state pediatric pneumonia admission rates.

METHODS. We performed retrospective analyses of data for all Pennsylvania-resident children 2 months through 17 years of age who were admitted to acute care hospitals with a principal diagnosis of pneumonia in 2003 or 2004. We divided the admissions into 2 groups (all pneumonia and pneumonia excluding coded comorbid chronic conditions) and calculated admission rates for each Pennsylvania county.

RESULTS. There were 5429 pediatric pneumonia admissions during the 12-month study period, of which 4948 (91.1%) were included in the study. The Pennsylvania state admission rate for all pneumonia was 156.3 admissions per 100000 children. County admission rates for all pneumonia ranged from 77.0 admissions per 100000 children to 457.6 admissions per 100000 children. Similar geographic patterns were seen among the 2851 admissions that remained in the second group after the exclusion of 2097 records (42.4%) coded for comorbid chronic conditions. The Pennsylvania state admission rate for pneumonia without chronic conditions was 90.0 admissions per 100000 children. County admission rates for pneumonia without comorbid chronic conditions ranged from 18.3 admissions per 100000 children to 350.3 admissions per 100000 children. Sixty-two (93%) of 67 counties remained in the same or an adjacent admission rate quintile after children with comorbid chronic conditions were excluded. On average, the county admission rates for pneumonia without comorbid chronic conditions were 58.1% of their admission rates for all pneumonia.

Conclusions. County pediatric pneumonia admission rates vary widely, even among geographically contiguous and demographically similar counties. Excluding children with comorbid chronic conditions, to control for varying community disease burdens, did not alter substantially the county rank order or the pattern or degree of variations in admission rates in our study.

Key Words: infant • child • preschool • adolescent • pneumonia • small-area analysis • utilization

Abbreviations: ICD-9-CM—International Classification of Diseases, Ninth Revision, Clinical Modification

FOR MORE THAN 3 decades, a number of authors1–4 have used small-area analyses to describe geographic variations in health services use. The Dartmouth Atlas of Health Care by Wennberg and Cooper5 demonstrated graphically the wide variations seen in the use of services by clinically similar populations. Fine et al6 documented these variations in community-acquired pneumonia among adults (Diagnosis Related Group 89). Although there have been exceptions,7,8 the bulk of the previous research has centered on adult populations. In this study, we sought to compare county rates of hospital admissions in a large, geographically and socioeconomically diverse state for a common pediatric condition, namely, community-acquired pneumonia. To identify variations potentially arising from geographic clustering of chronic health conditions, we also sought to assess the contribution of chronic conditions to county and state pediatric pneumonia admission rates.


   METHODS
TOP
 ABSTRACT
 METHODS
RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Hospitals in Pennsylvania are required to submit data to the Pennsylvania Health Care Cost Containment Council for each inpatient discharge. These data include demographic data such as patient residence, age, and gender; diagnosis and procedure data; and financial data such as total charges and payer type. The database and data collection system are described fully elsewhere.9 We performed retrospective analyses of a Pennsylvania statewide inpatient data set containing administrative data on all Pennsylvania-resident children 2 months through 17 years of age who were admitted to acute care hospitals with a principal diagnosis of pneumonia (within Diagnosis Related Group 081 or 091) during the period April 1, 2003 through March 31, 2004. We calculated pediatric pneumonia admission rates per 100000 children in the population for each Pennsylvania county. Admissions were assigned to a particular county on the basis of the child's county of residence. We excluded transfers from other hospitals to eliminate double counting; each pneumonia episode was counted only once. We obtained county child population demographic data from US Bureau of Census data and county Medicaid enrollment data from Pennsylvania Department of Health data. We adjusted raw county pneumonia admission rates by using indirect standardization to account for differences in the mean age (in months), percentage of male subjects, percentage of black subjects, and percentage of subjects with Medicaid coverage.

To assess the potential impact of comorbid chronic conditions on county pediatric pneumonia admissions rates, we divided the admissions into 2 groups, ie, all pneumonia and pneumonia excluding coded comorbid chronic conditions. Adjusted county admission rates were calculated for both groups by using the same total pediatric population in the denominator, because population numbers of residents with chronic conditions were not available. We construed chronic conditions broadly, to maximize the sensitivity of comorbidity case ascertainment. Table 1 summarizes the categories, International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) codes, and frequency counts of the chronic conditions we observed. We counted a chronic condition as comorbid if it was coded in any secondary diagnosis position. Each record was examined sequentially for the presence of a chronic condition ICD-9-CM code in the first through eighth secondary positions. The frequency counts in Table 1 reflect the most prominently positioned chronic condition coded in each record. As soon as the analysis encountered a chronic condition code, the record was flagged and categorized. No attempt was made to identify multiple comorbidities. A full listing of the ICD-9-CM codes we observed is available on request. We then calculated descriptive statistics for the patient ages (in months) and the total hospital charges for the pneumonia admissions without a coded chronic condition and those with a coded chronic condition present and compared them with t tests (pooled and Satterthwaite), to confirm that we had defined discrete populations effectively.


View this table:
[in this window]
[in a new window]
  		TABLE 1 Frequency of Most Prominently Coded Comorbid Chronic Conditions for Pennsylvania Pediatric Pneumonia Admissions (April 1, 2003 Through March 31, 2004)

 

   RESULTS
TOP
 ABSTRACT
 METHODS
 RESULTS
DISCUSSION
 CONCLUSIONS
 REFERENCES
 
There were 5429 pediatric pneumonia admissions to Pennsylvania hospitals during the 12-month study period, of which we included 4948 (91.1%) in the study. We excluded admissions for infants <2 months of age, non-Pennsylvania residents, and transfers from other acute care hospitals. We identified 2097 study admissions (42.4%) with coded comorbid chronic conditions. We assume that the remaining 2851 admissions for pneumonia excluding chronic conditions represent community-acquired pneumonia among nonimmunocompromised hosts, but we cannot assess the impact of undercoding of comorbidity in the study data set. The means, medians, and interquartile ranges for ages (in months) and total hospital charges for the various study populations are shown in Table 2.


View this table:
[in this window]
[in a new window]
  		TABLE 2 Ages and Total Hospital Charges for Pennsylvania Pneumonia Admissions for Children With and Without Comorbid Chronic Conditions (April 1, 2003 Through March 31, 2004)

 
The Pennsylvania state admission rate for all pneumonia was 156.3 admissions per 100000 children. Adjusted county admission rates are shown in Fig 1. There was substantial county-to-county variation. County admission rates for all pneumonia ranged from 77.0 admissions per 100000 children to 457.6 admissions per 100000 children. Five of 6 regional health districts had ≥1 county in the highest admission rate quintile; 4 of 6 had ≥1 county in the lowest quintile. Six of 13 highest-quintile counties were geographically contiguous with ≥1 lowest-quintile county. Pennsylvania's 2 largest urban counties fell in the middle quintile, as did several sparsely populated rural counties.


Figure 1
View larger version (49K):
[in this window]
[in a new window]
  		FIGURE 1 All patients with pneumonia. Pennsylvania county rates of pediatric pneumonia hospital admissions per 100 000 population of age 2 months through 17 years (April 1, 2003 through March 31, 2004) are shown.

 
Similar geographic patterns were seen among the 2851 admissions that remained in the second group after the exclusion of records coded for comorbid chronic conditions. The Pennsylvania state admission rate for pneumonia without comorbid chronic conditions was 90.0 admissions per 100000 children. Adjusted county admission rates for pneumonia without chronic conditions ranged from 18.3 admissions per 100000 children to 350.3 admissions per 100000 children (Fig 2). Sixty-two (93%) of 67 counties remained in the same or an adjacent admission rate quintile after children with chronic conditions were excluded. On average, the county admission rates for pneumonia without chronic conditions were 58.1% of their admission rates for all pneumonia (95% confidence interval: 31.5–84.6%).


Figure 2
View larger version (49K):
[in this window]
[in a new window]
  		FIGURE 2 Patients with pneumonia without comorbid chronic conditions. Pennsylvania county rates of pediatric pneumonia hospital admissions per 100 000 population of age 2 months through 17 years (April 1, 2003 through March 31, 2004) are shown.

 
Because of the small numbers of admissions for some counties, we elected to replicate our analyses for the previous year, 2002, to test the stability of the findings. Although the 2002 state admission rate for all pneumonia was somewhat less at 132.2 admissions per 100000 children, compared with 156.3 admissions per 100000 children, the adjusted county rates continued to show wide variability, ranging from 52.7 admissions per 100000 children to 233.0 admissions per 100000 children. The high and low counties remained the same, and 53 (80.6%) of 67 counties maintained their rankings in the same or an adjacent quintile. The pattern and distribution of high and low adjusted county admission rates remained quite consistent with the findings presented above.


   DISCUSSION
TOP
 ABSTRACT
 METHODS
 RESULTS
 DISCUSSION
CONCLUSIONS
 REFERENCES
 
Our data demonstrate that the geographic variations observed previously for adult health service utilization also occur among children in Pennsylvania with pneumonia. Although we identified nearly one half of all pneumonia admissions as being associated with comorbid chronic conditions, elimination of these cases with comorbid chronic conditions did not materially change the degree and pattern of observed variations. Although in a state agency we are quite familiar with access and referral patterns across our state, we were unable to identify a plausible explanation for our observations beyond prevailing community practice patterns. Although reproducibility of the pattern of variation across multiple years is consistent with the prevailing community practice hypothesis, our data did not permit us to assess the impact of localized epidemics on county pediatric pneumonia admission rates.

These observations are of concern for several reasons. Hospital admission is a frightening, intrusive, and disruptive intervention for young children and their families. Evidence suggests that, like adults, children are at significant risk for medical errors in the inpatient setting.10 In 2003, the direct costs of these 4948 Pennsylvania hospitalizations (with the assumption of the statewide ratio of charges to net patient revenue of 3.29:1)11 were more than $15 million. Although the data set did not permit rigorous payer analysis, the limited available data and past experience suggest that much of this expense was born by the state Medicaid program and the State Children's Health Insurance Program. Indirect costs related to lost parental workdays, increased child care expenses for other siblings, and decreased employee productivity are also presumably substantial.

Our data do not permit us to determine the "right" county rate of admissions for treatment of pediatric pneumonia, with or without comorbid chronic conditions. Some counties with high admission rates may confront risk factors for which our analysis could not control adequately. Some Pennsylvania counties with low rates may face inadequate access, although Montour County, a small rural county in central Pennsylvania that was in the lowest quintile, is home to Geisinger Medical Center, a nationally recognized academic health system. Although we were unable to assess community pneumonia prevalence in the manner described by Shwartz et al,12 the reproducibility of our findings over time leaves us confident that unidentified patient risks or variability in access, either alone or in combination, do not explain sufficiently the high degree of variability we observed.

Our data reflect only the pediatric pneumonia experience in Pennsylvania in 2003 and 2004, and additional research is needed to determine whether our findings related to pneumonia can be generalized to other pediatric disease states and to other geographic areas. We think that replication of our findings elsewhere would argue in favor of an authoritative national guideline for pediatric pneumonia admission standards that could support reductions in the small-area variations observed in our study.


   CONCLUSIONS
TOP
 ABSTRACT
 METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
REFERENCES
 
County pediatric pneumonia admission rates vary widely, even among geographically contiguous and demographically similar counties. Excluding children with comorbid chronic conditions, to control for varying community disease burdens, did not alter substantially the county rank order or the pattern or degree of variation in admission rates in our study. Hospitalization is an intrusive intervention that has substantial physical and emotional effects on children and their families, with significant costs to public and private payers. Researchers, policymakers, and decision-makers should explore the sources of unnecessary variations in pediatric pneumonia admission rates and seek mechanisms to eliminate them.


   ACKNOWLEDGMENTS
 
This study was supported by an internal preexisting budget of the Pennsylvania Health Care Cost Containment Council that is allocated for special projects.

We acknowledge and appreciate the data analysis contributions of Edward Hain, BA.


   FOOTNOTES
 
Accepted Jul 29, 2005.

Address correspondence to Christopher P. Gorton, MD, MHSA, 810 Zermatt Dr, Hummelstown, PA 17036. E-mail: kitgorton{at}comcast.net

Dr Gorton, as principal investigator of this study, had full access to all of the data in this study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Authors from the Pennsylvania Health Care Cost Containment Council participated in the design and conduct of the study, in the analysis and interpretation of the data, and in the preparation, review, and approval of the manuscript. Data for the study were collected by the Pennsylvania Health Care Cost Containment Council.

The authors have indicated they have no financial relationships relevant to this article to disclose.


   REFERENCES
TOP
 ABSTRACT
 METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 

  1. Fisher ES, Wennberg JE, Stukel TA, Sharp SM. Hospital readmission rates for cohorts of Medicare beneficiaries in Boston and New Haven. N Engl J Med. 1994;331 :989 –995[Abstract/Free Full Text]
  2. Restuccia J, Shwartz M, Ash A, Payne S. High hospital admission rates and inappropriate care. Health Aff (Millwood). 1996;15 :156 –163[Abstract]
  3. Ashton CM, Petersen NJ, Souchek J, et al. Geographic variations in utilization rates in Veterans Affairs hospitals and clinics. N Engl J Med. 1999;340 :32 –39[Abstract/Free Full Text]
  4. Baicker K, Chandra A, Skinner JS, Wennberg JE. Who you are and where you live: how race and geography affect the treatment of Medicare beneficiaries. Health Aff (Millwood). 2004;(suppl web exclusive):VAR33–VAR44
  5. Wennberg JE, Cooper MM, eds. The Dartmouth Atlas of Health Care 1998. Chicago, IL: American Hospital Publishing;1998
  6. Fine MJ, Auble TE, Yealy DM, et al. A prediction rule to identify low-risk patients with community-acquired pneumonia. N Engl J Med. 1997;336 :243 –250[Abstract/Free Full Text]
  7. Kocevar VS, Bisgaard H, Jönsson L, et al. Variations in pediatric hospitalization rates and costs between and within Nordic countries. Chest. 2004;125 :1680 –1684[Abstract/Free Full Text]
  8. Chabra A, Chavez GF, Taylor D. Hospital use by pediatric patients: implications for change. Am J Prev Med. 1997;13 :30 –37[Medline]
  9. Pennsylvania Health Care Cost Containment Council. 2003 Hospital Performance Report: Technical Notes. Harrisburg, PA: Pennsylvania Health Care Cost Containment Council;2004
  10. Owens PL, Thomson J, Elixhauser A, Ryan K. Care of Children and Adolescents in US Hospitals. Rockville, MD: Agency for Healthcare Research and Quality;2003:22 . HCUP Fact Book 4, AHRQ Publication 04-0004
  11. Pennsylvania Health Care Cost Containment Council. 2003 Financial Analysis: Volume 1: General Acute Care Hospitals. Harrisburg, PA: Pennsylvania Health Care Cost Containment Council;2004
  12. Shwartz M, Peköz EA, Ash AS, et al. Do variations in disease prevalence limit the usefulness of population-based hospitalization rates for studying variations in hospital admissions? Med Care. 2004;43 :4 –11
PEDIATRICS (ISSN 1098-4275). ©2006 by the American Academy of Pediatrics




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow P3Rs: Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when P3Rs are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow E-mail this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My File Cabinet
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via CrossRef
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Gorton, C. P.
Right arrow Articles by Jones, J. L.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Gorton, C. P.
Right arrow Articles by Jones, J. L.
Related Collections
Right arrow Infectious Disease & Immunity