Hospitalists and improved cost savings in patients with bacterial pneumonia at a state level.

Objectives: In the hospitalist literature, most studies have focused on outcomes related to cost savings for individual hospital systems. This study sought to determine if hospitalists could improve cost savings at a state level.

Methods: This is a retrospective analysis of a statewide database for inpatients in 2002 with bacterial pneumonia. The primary outcomes measured were mean length of stay (LOS) and mean charges per patient between hospitalists and nonhospitalists. The secondary outcome measured was percentage of patients by severity of illness between the groups.

Results: The difference of LOS in the moderate illness category was 4.9 days for hospitalists and 5.2 for nonhospitalists (P = 0.04). The major illness category was 7.4 and 8 (P = 0.03), and the extreme illness category was 10.6 and 12.9 (P = 0.02). The difference of mean charges per patient in the major category were $20,950 and $23,259 (P = 0.03) and $42,045 and $56,867, respectively (P = 0.002), in the extreme category. Patients in the major/extreme categories of illness accounted for 41% of hospitalist patients versus 32% of nonhospitalist patients (P < 0.001).

Conclusions: Hospitalists have shorter LOS, lower charges per patient, and admit a larger proportion of high acuity patients at a state level.

Key Words: community-acquired pneumonia, cost per patient, hospitalist, length of stay, South Carolina 

**********

Hospitalists are defined by the Society of Hospital Medicine as "clinicians whose primary professional focus is the general medical care of hospitalized patients." The hospitalist movement has been growing exponentially since its inception less than a decade ago. The primary forces behind the movement involve logistical and financial issues relating to the time and cost pressures on hospitals, physicians, and managed care groups. These pressures, combined with increasing acuity of hospitalized patients and the need for accelerated pace of hospital stays, has led to the development of the hospitalist as a specialty. The state of South Carolina has also experienced a rapid growth of hospitalist programs. According to the Society of Hospitalist Medicine, there are more than 80 members in the Society practicing in the state of South Carolina, presumably with more practicing hospitalists that are not members.

Most of the studies of hospitalist programs to date have focused primarily on outcomes related to cost savings for hospital systems. It is well established in the literature that hospitalist programs can decrease the length of stay and cost per patient at individual medical centers. (1-14) Critics of hospitalist literature note that most studies have been done at individual medical centers with small numbers of hospitalists, as recently reported by Wachter. (15) No studies to date have evaluated the impact of hospitalist programs at a state or national level.

Given the potential impact that hospitalists can have on the cost and quality of health care for a community, we sought to determine if hospitalist programs in the state of South Carolina could improve outcomes in a manner similar to those at individual medical centers.

Materials and Methods

This study sought to determine whether hospitalists practicing in the state of South Carolina have better patient outcomes than nonhospitalist physicians taking care of patients with bacterial pneumonia. We also sought to determine whether hospitalists in the state assume care of higher acuity patients than nonhospitalists. Our study was a retrospective analysis of a statewide database. The South Carolina Office of Research and Statistics (SCORS) was consulted to provide information regarding patient outcomes. The Health and Demographics Section of SCORS collects, distributes, and interprets health and demographic data in the state of South Carolina. Hospitals are mandated by law to submit data to SCORS on a quarterly basis. The time line for submission is that 95% of the data must be submitted within 45 days after the close of the quarter and the remaining 5% must be submitted within 45 days from the close of the following quarter. The data are run through edits and required to be 99.5% accurate and 99% complete. Failure to comply with submission is subject to a civil fine of up to $10,000.

There was a total of 29 hospitals included in the analysis, which included all short-term acute care hospitals in the state of South Carolina with hospitalists on staff (excluding military, VA, and specialty hospitals). Those hospitals without hospitalists on staff were excluded to ensure an appropriate comparison group. Institutional review board approval was obtained from The Medical University of South Carolina.

The primary outcomes measured included mean length of stay and mean charges per patient for those admitted to South Carolina hospitals with a primary diagnosis of bacterial pneumonia. ICD-9 codes were used to identify cases. Only those cases in which the condition was the primary diagnosis were included. Pneumonia ICD-9 codes included 481.XX, 482.2X, 482.3X, 482.9X, 483.XX, 485.XX, and 486.XX. The secondary outcome measured was the percentage of patients cared for by each group, stratified by severity of illness (mild, moderate, major, and extreme). The levels of severity were determined by assessing the patient data by APR-DRG software, which assigns each record a severity category. This software is a widely used adjustment method used by 25 state and federal agencies including the Agency for Health Care Research and Quality. It adjusts the primary diagnosis for severity and takes into account age, comorbidities, and procedures.

Hospitalist physicians were identified through the Society of Hospitalist Medicine, which accounts for only those hospitalists registered through the society. Their South Carolina license numbers were used to determine the patients for which they were the discharge attending. There were 53 hospitalist physicians and 1,489 nonhospitalist physicians who admitted patients with bacterial pneumonia in the year 2002. All data for the year 2002 were included.

The primary outcomes of mean length of stay (LOS) and mean charges per patient between hospitalists and nonhospitalists were compared by a simple t test. The secondary outcome of percentage of patients in each severity of illness category between hospitalists and nonhospitalists was compared by a [chi square] trend. Values of P less than 0.05 were considered significant.

Results

There were 11,969 patients admitted to these South Carolina hospitals in 2002 with the primary diagnosis of bacterial pneumonia. Approximately 10% of the patients (n = 1,214) were treated by hospitalist physicians, and 90% (n = 10,755) were treated by nonhospitalist physicians. Hospitalists and nonhospitalists were similar in age, years since medical school graduation, and sex (Table 1).

The difference of mean LOS in the minor severity category was similar between groups. In the moderate category, LOS was 4.9 days for hospitalists and 5.2 for nonhospitalists (P = 0.04); for the major category, LOS was 7.4 and 8, respectively (P = 0.03); and for the extreme category, LOS was 10.6 and 12.9, respectively (P = 0.02) (Table 2).

The difference of mean charges per patient in the minor and moderate categories was not statistically significant. In the major category, mean charges were $20,950 and $23,259 for the hospitalist and nonhospitalist groups, respectively (P = 0.03); for the extreme category, they were $42,045 and $56,867, respectively (P = 0.002) (Table 3).

Those patients in the major/extreme categories of illness accounted for 41% of hospitalist patients compared with only 32% of nonhospitalist patients ([[chi square].sup.trend] = 80.59, P < 0.001; Table 4).

Discussion

The growth of hospitalist programs throughout the nation has been accompanied by the growth of literature supporting their role in improving cost savings for individual hospital systems. This literature has predominately originated from academic medical centers with established hospitalist programs. This study combined all registered hospitalists across the state, regardless of duration of existence of their hospitalist program or their affiliation with an academic medical center. There was no significant difference between the nonhospitalist and hospitalist physicians with respect to age, sex, or years since medical school graduation, implying that this was an appropriate comparison group. In addition, only those hospitals with hospitalists on staff were chosen to study to ensure this appropriate comparison group.

Bacterial pneumonia was chosen, as it is one of the top three indications for inpatient hospitalizations in the state of South Carolina (excluding labor and delivery). It accounted for 2.3% of all inpatient admissions in South Carolina for the year 2002. This study has shown that in this state, hospitalists care for a disproportionate number of severely affected patients with bacterial pneumonia. Hospitalists decrease the LOS and charges for these patients and therefore pass on a significant cost savings.

A significant limitation of the study is the difficulty in defining a hospitalist. Since there is no board certification for the specialty, hospitalists are defined by the perception of their primary responsibility, that is, the care of patients in the hospital. This study ascertained a hospitalist as one who is registered with the Society of Hospital Medicine. Since membership is elective, there are certain to be practicing hospitalists in the state that are not registered with the society. The result of not including these physicians in the hospitalist category, however, probably would lead to a dilution of the effect of the findings rather than an inflation. In addition, this study was restricted to only those patients with bacterial pneumonia, which limits its generalizability. However, similar studies could easily be undertaken to include a wider breadth of diagnoses across the state.

Conclusion

Hospitalists provide dedicated, full-time care to patients with the most common inpatient general medical conditions. We have shown that in patients with bacterial pneumonia, hospitalists care for those who are more severely affected, decrease these patients' LOS, and pass on a significant cost savings to the health care system. As the growth of hospitalist programs continues to expand, both cost savings and outcomes related to quality of care will need to be followed. These outcomes must be evaluated at the state and national levels to justify the continued expansion of these services.

Most folks are about as happy as they make up their minds to be.
--Abraham Lincoln

Table 1. Physician demographics

                     Hospitalist  Nonhospitalist
                     (n = 53)     (n = 1489)

Age                  42.2 years   44.9 years
% Male               83.02%       78.24%
Years since medical  14.1 years   17.8 years
  school graduation

Table 2. Mean length of stay in days (standard deviation)

Severity
of illness  Hospitalist  Nonhospitalist  P

Minor        3.5 (2.5)    3.1 (1.9)      0.0749
Moderate     4.9 (3.4)    5.2 (3.7)      0.0446
Major        7.4 (5.2)    8.0 (6.1)      0.0349
Extreme     10.6 (8.2)   12.9 (11.3)     0.0191

Table 3. Mean charges per patient in dollars (standard deviation)

Severity
of illness  Hospitalist       Nonhospitalist    P

Minor        $7,906 (5,067)    $7,018 (5,743)   0.0561
Moderate    $12,124 (9,745)   $12,503 (9,517)   0.3704
Major       $20,950 (18,624)  $23,259 (24,790)  0.0266
Extreme     $42,045 (37,112)  $56,867 (66,299)  0.0019

Table 4. Patients by category of illness ([chi square] = 80.59, P <
0.001)
          Hospitalist  Nonhospitalist
          (n = 1214)   (n = 10,755)

Minor     130 (10.7%)  2211 (20.6%)
Moderate  590 (48.6%)  5117 (47.6%)
Major     402 (33.1%)  2820 (26.2%)
Extreme    92 (7.6%)    607 (5.6%)

Accepted December 4, 2004.

References

1. Wachter RM, Goldman L. The hospitalist movement 5 years later. JAMA 2002;287:487-494.

2. Wachter RM, Katz P, Showstack J. Reorganizing an academic medical service: impact on cost, quality, patient satisfaction, and education. JAMA 1998;279:1560-1565.

3. Diamond HS, Goldberg E, Janosky JE. The effect of full-time faculty hospitalists on the efficiency of care at a community teaching hospital. Ann Intern Med 1998;129:197-203.

4. Kroger E, Grant M. Are hospitalists truly more efficient? A natural experiment. J Gen Intern Med 1998;13:S66.

5. Stein MD, Hanson S, Tammaro D. Economic effects of community vs hospital-based faculty pneumonia care. J Gen Intern Med 1998;13:774-777.

6. Craig DE, Hartka L, Likosky WH, Implementation of a hospitalist system in a large health maintenance organization: the Kaiser Permanente experience. Ann Intern Med 1999; 130:355-359.

7. Rifkin WD, Conner D, Silver H. Comparison of processes and outcomes of pneumonia care between hospitalists and community-based primary care physicians. Mayo Clin Proc 2002;77:1053-1058.

8. Davis KM, Koch KE, Harvey JK. Effects of hospitalists on cost, outcomes, and patient satisfaction in a rural health system. Am J Med 2000;108:621-626.

9. Halpert AP, Pearson SD, LeWine HE. The impact of an inpatient physician program on quality, utilization, and satisfaction. Am J Manag Care 2000;6:549-555.

10. Bellet PS, Whitaker RC. Evaluation of a pediatric hospitalist service: impact on length of stay and hospital charges. Pediatrics 2000;105:478-484.

11. Auerbach AD, Wachter RM, Katz. P. Implementation of a voluntary hospitalist service at a community teaching hospital: improved clinical efficiency and patient outcomes. Ann Intern Med 2002;137:859-865.

12. Hackner D, Tu G, Braunstein GD. The value of a hospitalist service. Chest 2001;119:580-589.

13. Palmer HC, Armistead NS, Elnicki DM. The effect of a hospitalist service with nurse discharge planner of patient care in an academic teaching hospital. Am J Med 2001;111:627-632.

14. Meltzer D, Manning WG, Morrison J. Effects of physician experience on costs and outcomes on an academic general medicine service: results of a trial of hospitalists. Ann Intern Med 2002; 137:866-874.

15. Wachter, RM. Hospitalists in the United States: mission accomplished or work in progress? N Engl J Med 2004;350:1935-1936.

RELATED ARTICLE: Key Points

* The number of hospitalists in the country is growing exponentially.

* Hospitalists improve cost outcomes in patients with community-acquired pneumonia at a state level.

* Hospitalists care for a larger proportion of more severely ill patients than nonhospitalist physicians.

Danielle B. Scheurer, MD, Justin G. Miller, MD, Dwight I. Blair, MD, Pam J. Pride, MD, Gena M. Walker, MD, and Patrick J. Cawley, MD

From the Department of Internal Medicine, Hospitalist Program, Medical University of South Carolina, Charleston, SC.

Institutional review board approval was obtained from the Medical University of South Carolina.

The authors received no financial compensation and had no commercial interests related to the contents of this article.

Reprint requests to Dr. Danielle B. Scheurer, Medical University of South Carolina, Department of Internal Medicine, Hospitalist Program, 171 Ashley Avenue, Charleston, SC 29425. Email: [e-mail]