 |
|
 |
|
 |
 |
|
 |
 |
 |
 |
 |
 |
 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
* Department of Medicine, Children’s Hospital, Harvard Medical School, Boston, Massachusetts
Department of Pediatrics, University of Utah Health Sciences Center, Salt Lake City, Utah
Department of Ambulatory Care and Prevention, Harvard Medical School and Harvard Pilgrim Health Care, Boston, Massachusetts
|| Harvard Vanguard Medical Associates, Boston, Massachusetts
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Design. Interrupted time-series study of general pediatric patients admitted to a freestanding pediatric teaching hospital from 1993 to 1998. The intervention group consisted of all patients admitted to a staff model not-for-profit HMO that began using hospitalists in October 1996. Patients in other HMOs and traditional insurance groups were studied for comparison. The main outcomes were mean LOS and inflation-adjusted costs. Other outcomes included parents’ ratings of care and mortality, readmission, and follow-up rates.
Results. Immediately after the introduction of the hospitalist system, mean LOS for staff model not-for-profit HMO fell 12% (0.3 days), and mean cost of hospitalization decreased 16% ($217) compared with prehospitalist levels. Parental ratings of care, initially somewhat low, improved substantially. Seven-day follow-up rates, mortality, and readmission rates did not change. Comparison groups experienced no concurrent improvements in LOS, cost, parental ratings, mortality, or readmission rates.
Conclusions. A pediatric hospitalist system within a staff-model HMO significantly improved LOS, cost, and parental ratings of care without affecting rates of posthospitalization follow-up. Additional studies are needed both to measure broader aspects of quality and to assess the impact of such a program on patients being cared for through other types of delivery and insurance systems.
Key Words: hospitalists • pediatrics • time-series analysis • health services research
Abbreviations: LOS, length of stay • SMHMO, staff-model health maintenance organization • HMO, health maintenance organization • FTE, full-time equivalencies • APDRG, All Patient Diagnosis-Related Groups
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
In a survey of pediatric department chairs in the United States and Canada, we found that 50% of academic pediatric centers have begun to use hospitalists, and another 27% are planning to use them.10 Studies of pediatric hospitalist systems are needed to better understand the value of hospitalists in pediatrics. Data from internal medicine hospitalist systems are not directly pertinent, because children’s diseases are different, LOS is shorter, the cost of hospitalization is lower, and the structure and finance of pediatric primary care is different.11 A system of care that decreases cost and LOS for adults might not do so for children.
Published studies have reached contradictory conclusions about the impact of pediatric hospitalists. A retrospective study at Southern California Children’s Hospital of 722 children admitted with asthma or bronchiolitis found no differences in costs, LOS, or clinical outcomes between patients cared for by hospitalists and those cared for by community pediatricians.12 However, potential for selection bias was introduced by differences in the proportion of patients on each service with MediCal (Medicaid) insurance, and differences in the proportion admitted through the emergency department. In addition, the small size of the study limited the power to detect potentially important differences of 10% to 15%.
In contrast, mean LOS decreased by 0.3 days (11%; P = .005) and mean hospitalization charges decreased by $282 (9%; P = .01) after the introduction of hospitalists in a retrospective observational study at Children’s Hospital of Cincinnati.13 However, the 10-day readmission rate increased substantially from 1% to 3% (P = .006). The before-after design of the Cincinnati study limits the degree to which these changes can be attributed conclusively to the hospitalist system. For example, the changes may have been attributable to a co-intervention, year-to-year variation, or secular trend.
Introduction of a general pediatric hospitalist system by a not-for-profit staff-model health maintenance organization (SMHMO) in October 1996 provided us with an opportunity to determine the impact of hospitalists on the efficiency and quality of care in an academic pediatric referral center. Other general pediatric patients, including those admitted to a similar health maintenance organization (HMO2) that did not use hospitalists, served as contemporaneous comparison groups. We used interrupted time-series analysis, a regression technique that controls for preintervention trends,14–17 to evaluate the impact of hospitalists on LOS, costs, and resource utilization, while controlling for secular trend and other changes in the health care system during the study period. We also measured changes in mortality, readmission rates, continuity of care, and parent ratings of care before and after the introduction of the hospitalist system.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Planning and implementation of the hospitalist program took place over a 10-month period, beginning in January 1996. Specifically, a pediatric leadership group at SMHMO met in January 1996 to discuss inefficiencies in the existing traditional rounding system (eg, amount of time required for pediatricians to travel to and from the hospital, and outpatient revenue lost during this travel time and time spent in the hospital). In response to these issues, the possibility of establishing a hospitalist system for all SMHMO patients was raised. In March 1996, as the leadership group decided to move forward with the system change, a survey was administered to all SMHMO pediatricians that discussed existing inefficiencies and asked for their response to a tentative proposal to establish a hospitalist system. Over the ensuing 7 months, plans were finalized and hospitalists took over the care of all SMHMO inpatients in October 1996.
SMHMO admits 
700 to 1000 general pediatrics patients per year, and typically has a census of 10 to 20 patients. Since October 1996, the patient census has been divided among 2 or 3 hospitalists, who spend 0.5 to 1.0 full-time equivalencies (FTEs) caring for inpatients. Housestaff and medical students are involved in the care of these patients and are supervised by the hospitalists. The nature of the relationship between staff physicians from SMHMO and trainees did not change with the advent of the hospitalist system and was fundamentally similar to the relationship between trainees and providers from other HMOs and private practices. On weekends, a pediatrician from the hospitalist group or from one of the local health centers cares for the patients.
All of the hospitalists hired by SMHMO have been experienced general pediatricians from the community. Individuals in the group working <1.0 FTEs as hospitalists spend the remainder of their time in administration, including management and utilization review. The hospitalist group employs nurse liaisons who aid in the care and discharge planning of all patients. There was little turnover in the hospitalist staff between 1993 and 1998; 2 of the 3 hospitalists working in 1998 had been there throughout the entire time period.
Comparison Groups
We studied 2 HMO comparison groups and 3 non-HMO comparison groups. HMO2 is a large, nonstaff model, not-for-profit HMO that admits 400 to 500 patients to the study hospital per year. It has never used hospitalists. All HMO2 patients admitted during the study period were cared for by their own pediatricians or by cross-covering pediatricians from their practices.
N-SMHMO is a nonstaff model, not-for-profit HMO that is indirectly affiliated with SMHMO. Most of these patients were cared for by their own pediatricians or their partners, but some were cared for by the same hospitalists caring for SMHMO patients from October 1996 through 1998.
We also studied all other patients admitted from 1993 to 1998 to the general pediatrics service with traditional indemnity insurance, Medicaid, or self-pay. Hospitalists cared for none of these patients. These patients were less similar to our main study group than the HMO comparison groups, but we chose to examine them as well to more completely understand global trends in pediatric care at the hospital during the study period.
Data Collection
We identified all patients 0 to 17 years of age discharged from the general pediatrics service from October 1993 to June 1998; surgical and subspecialty patients were not included, nor were 23-hour observation patients. We extracted demographic, case-mix, financial, and clinical data from the hospital database. To validate insurance status, SMHMO patients identified through the hospital’s computerized database were cross-referenced with patient lists maintained by their insurer. The small number of patients admitted from SMHMO’s 3 outlying centers were excluded from the study (n = 206), as hospitalists had been caring for this group of patients before the widespread implementation of the hospitalist system in October 1996.
Cost and resource outcomes included LOS, total inflation-adjusted direct costs, laboratory costs, radiology costs, and pharmacy charges. Costs were determined using the hospital’s utilization-based cost-allocation system. This system measures costs of services provided, controlling for relative shifts in charges over time, and for differences in negotiated payment scales by different payers. Because the hospital had not applied its cost-allocation system to all pharmacy services, we looked at pharmacy charges as a proxy for cost. To adjust for inflation as precisely as possible, we adjusted costs for inflation using the Boston-Brockton-Nashua Medical Care Index of the Consumer Price Index.18 To confirm the validity of our findings, we subsequently readjusted for inflation using the more conservative all-cities general inflation rate from the Consumer Price Index, which produced no important differences. Costs are reported in 1993 dollars.
Clinical outcomes included rates of readmission within 2 weeks, mortality, and postdischarge follow-up rates. The readmission and mortality rates were extracted from the hospital’s database. Rates of follow-up visits to a doctor after discharge were obtained for SMHMO patients only for 1 year before and 1 year after the introduction of hospitalists in October 1996. Two-day and 7-day follow-up rates, as well as type of visit (unscheduled vs scheduled), were collected by merging hospital discharge data with physician encounter data from SMHMO’s database.
Data on parental ratings of overall hospitalization and key care processes were collected from the modified Picker survey of parents’ perceptions of clinical care.19 This telephone survey was administered 2 times per year from the autumn of 1995 through the autumn of 1997 (2 times before the hospitalist system and 3 times after). The parents of 600 hospitalized patients were interviewed during each survey period, within 2 weeks of discharge. We matched data from the Picker survey with our primary database to identify the parents of all general pediatrics patients interviewed. We selected 12 items for analysis a priori that were of particular interest in evaluating hospitalist and traditional inpatient care systems, including parents’ overall rating of hospitalization and ratings of important inpatient care processes. Examples included parents’ ratings of physicians’ availability and communication, and parents’ confidence and trust in physicians in the hospital.
Analyses
Demographic and case-mix differences between the intervention and comparison groups were analyzed using Wilcoxon tests for nonnormal continuous variables and 
2 tests for dichotomous variables. Mean quarterly LOS and costs were calculated for each group. Outlying costs and LOS (>3 standard deviations above the mean) were truncated to 3 standard deviations above the mean, a standard established in previous studies of hospitalist systems.5,13
Serial interrupted time-series models were built for quarterly LOS and costs for the intervention group, and for each of our comparison groups. Changes in level and trend before and after October 1996 were analyzed, with baseline slope variables included to control for secular trend, as is typically done in time-series analyses. Inclusion of this variable, which detects upward or downward trends in the baseline data, allows any systematic trend in the data beginning before the intervention (and thus unrelated to it) to be captured and controlled for in the model. Each model was adjusted for quarterly All Patient Diagnosis-Related Groups (APDRG) [NY8 version] case-mix to control for any shifts in severity of illness over time, and to control for differences in severity between groups.20 After a careful review of written documentation from 1996 that outlined the plans and implementation of the hospitalist system, we chose, before doing the analyses, to exclude the peri-interventional second and third quarters of 1996 from time series analyses. After the announcement to all pediatricians in March, the system change at SMHMO had effectively begun, but was not fully implemented until hospitalists took over all inpatient care in October. Exclusion of peri-interventional data points is typically done in time series analysis to address this common problem, as failure to do so may result in overestimation or underestimation of effect size.16,21 To make the analyses of other groups comparable, the second and third quarters of 1996 were excluded for them as well.
Because of insufficient case numbers and/or insufficient numbers of time-ordered data points, mortality, readmission rates, and follow-up rates could not be reliably analyzed by time-series. These outcomes were instead analyzed for each group pre- versus post-October 1996 using 2 tests. Simple logistic regression models controlling for demographic variables and severity of illness were also built for readmission rate and mortality.
Parents’ ratings of inpatient care pre- versus post-October 1996 were analyzed by 2 tests. Ratings’ of each question with >2 ordinal responses were dichotomized into best possible response versus other. Because the number of respondents in some comparison groups was quite limited, we combined the 2 HMO comparison groups into 1 group and the 3 non-HMO comparison groups into another for this analysis.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
2.5 days, among the lowest of the groups studied. There was no statistically significant trend up or down at baseline. Despite a low baseline LOS, mean LOS fell 0.3 days (P = .005) to 2.2 days after the introduction of the hospitalist system. After this initial drop, there was no statistically significant trend up or down in mean LOS.
Comparison Groups
We performed separate time-series analyses of HMO2, N-SMHMO, and the 3 non-HMO comparison groups, to see whether LOS also dropped in these groups after October 1996.
Time-series analysis of HMO2’s LOS revealed a borderline downward trend at baseline (-0.04 days per quarter; P = .07), but no significant change in level at the time of the intervention. There was a statistically significant trend toward increasing LOS from the end of 1996 to 1998 (0.13 days per quarter; P = .02).
Like HMO2, N-SMHMO had a downward trending LOS at baseline (-0.06 days per quarter; P = .002), demonstrated no drop in LOS level after October of 1996, and had a statistically significant trend upward in LOS from 1996 to 1998 (0.12 days per quarter; P = .009).
The time-series curves for SMHMO, HMO2, and N-SMHMO are displayed in Fig 1. The preintervention lines were similar, although SMHMO’s baseline LOS was slightly lower than the comparison groups’. The curves diverged in October of 1996, when SMHMO’s LOS dropped to a fairly steady 2.2 days, whereas the comparison groups’ climbed.
|
|
Although there was some tendency toward decreased laboratory costs and pharmacy charges after the implementation of the hospitalist system, time-series analyses of radiology costs, laboratory costs, and pharmacy charges showed no statistically significant changes (data not shown).
Comparison Groups
HMO2 experienced no drop in total hospitalization cost at the time of the intervention. There was a downward baseline trend (-$56 per quarter; P = .009) and a significant trend back upward from 1996–1998 ($90 per quarter; P = .03). N-SMHMO similarly had a baseline downward trend in cost (-$50 per quarter; P = .001), no drop at the time of the intervention, and a trend back upward from 1996–1998 ($65 per quarter; P = .04). The time-series curves for total hospitalization costs for SMHMO, HMO2, and N-SMHMO are shown in Fig 3. As was the case for LOS, the preintervention curves were similar, but increasingly diverged after October of 1996.
|
|
Clinical Outcomes
There were no deaths among hospitalized SMHMO patients on the general pediatrics service from 1993 to 1998. Readmission rates did not differ significantly before and after October 1996 by univariate analysis or by logistic regression controlling for APDRG case-mix and demographic variables. There were likewise no changes in mortality or readmission rates for any of the comparison groups.
Parental Ratings of Care
Surveys were completed by 377 parents of general pediatrics patients (92 parents of SMHMO patients). SMHMO parents’ rating of 3 items, including the overall rating of the hospitalization were significantly better after implementation of the hospitalist system. Although none of the differences in scores for the remaining 9 items were statistically significant, 8 of 9 items demonstrated improvement after the introduction of the hospitalist program.
Parental ratings of care before and after October 1996 were unchanged for comparison groups. No items were rated significantly differently for HMO or non-HMO comparison groups. For HMO comparison groups, 4 items improved after October 1996, and 8 worsened. For non-HMO comparison groups, 3 improved and 9 worsened (Table 2).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Time-series analysis is a powerful regression technique that can be used to evaluate the impact of an intervention when sufficient data are available over time. Unlike simple before-after analyses, time-series analysis can precisely identify the time that a change occurs, and can account for secular trend. Peri-interventional data points can be dropped from the analysis to factor out anticipatory changes just before an intervention. Unless a co-intervention occurs at precisely the same time as the intervention of interest, observed changes can be attributed to the intervention of interest with a high degree of confidence.
It is unclear why costs trended upward in SMHMO and the HMO comparison groups beginning in 1996, and why LOS trended upward for the comparison groups. Our case-mix measures were relatively crude20 and may have failed to detect admission of sicker, more complex patients later in the study period. In addition, we may have failed to detect increased availability and use of more expensive diagnostic tests or therapeutic agents. Alternatively, although unlikely, it is possible that there were unmeasured efforts in all groups to reduce LOS and costs early in the study which resulted in the baseline trend down, with a return back to the mean toward the latter years of the study as these efforts waned, a pattern less pronounced in SMHMO. Whatever the explanation for shifting trends, however, there was an immediate 12% to 16% benefit in LOS and costs respectively for SMHMO after the hospitalist system’s commencement, and an attenuated increase in costs (and no increase in LOS) for SMHMO than for the comparison groups in the latter years of the study. This suggests that after its initial benefit, the hospitalist system may have successfully countered pressures tending to increase costs and LOS across groups over this time period.
It is likely that the decrease in hospitalization costs seen for SMHMO was tied primarily to the decrease in LOS. LOS is a strong driver of total hospital costs, and the lack of significant changes in laboratory costs, radiology costs, and pharmacy charges for SMHMO would suggest that the alternative—decreased utilization of resources—was not the principal explanation for the decrease in total costs. However, there was some tendency toward decreased laboratory costs and pharmacy charges for SMHMO after introduction of the hospitalist system. The possibility that decreased utilization of resources explained part of the decrease in total costs merits exploration in future studies.
This study provides important pediatric data on the impact of hospitalists in settings where HMOs already are operating efficiently and have driven LOS down. At baseline, the LOS for SMHMO patients was among the lowest in the hospital at 2.5 days, but despite this, the hospitalist system provided substantial additional benefit. This study also suggests that hospitalists can have a sustained influence on efficiency of care. Improvements in LOS and costs persisted for 2 years after introduction of SMHMO’s hospitalist system.
This study also provides some data suggesting that increased efficiency need not come at the price of disruption in continuity of care or reductions in ratings of care.22 Immediate posthospitalization follow-up rates did not decrease with the introduction of hospitalists. Ratings of confidence and trust in physicians as well as communications regarding follow-up planning remained high. Overall ratings of hospital care improved significantly, as did families’ ratings of key aspects of care, including the consistency of information provided and the amount of time physicians spent discussing postdischarge care with them.
Although the improvements in parental ratings reflect well on SMHMO’s hospitalist system, they should be interpreted with caution. Before the advent of the hospitalist system, the 3 items that improved significantly had been rated lower than in comparison groups. Introduction of the hospitalist system may have corrected a problem in the processes of care specific to this group, and therefore, similar improvements may or may not be found in other hospitalist systems. Studies of parental ratings in other hospitalist systems will be needed to clarify this issue. In addition, it would be beneficial to study continuity of care in hospitalist systems unaffiliated with staff model HMOs. Probabilities of hospitalization (and most likely posthospitalization follow-up) may be affected by the strength of continuity of care relationships, and these in turn may vary by setting.23,24
We can only speculate about the specific aspects of the hospitalist system responsible for the documented improvements. It was the intention of SMHMO in making the change to a hospitalist system to improve the efficiency of care and satisfaction of patients and providers without harming quality; it is possible that formulation of this goal alone was the most important element in improving efficiency and patient satisfaction. It is possible that the improvements were instead more closely tied to the backgrounds of the hospitalists themselves. Their extensive inpatient experience, particularly their familiarity with the study hospital, coupled with their previous experience as community pediatricians may have contributed to their success. In addition, although the hospitalists received no direct financial incentives, they were aware that SMHMO’s intent in implementing a hospitalist system was to increase the efficiency of care and satisfaction of providers and patients, and this may have affected their performance. SMHMO liaison nurses were very active in the general pediatric wards, and they may have played an important, unmeasured role in facilitating timely discharge.
We principally compared the SMHMO hospitalist system with a previous cross-coverage system in which pediatricians from multiple large health centers rounded on their partners’ patients as well as their own. Although such cross-coverage arrangements are common, we did not compare the care given by hospitalists with that given by a patient’s specific primary care provider.
We did not perform a cost-effectiveness analysis of hospitalists, because we measured only direct inpatient costs related to patient care. A cost-effectiveness study would have to capture startup costs of the hospitalist system, administrative costs, and salaries; outpatient revenues generated by pediatricians during time freed from hospital responsibilities; and the short- and long-run distribution of benefits in hospitalist systems,25 but such analyses were beyond the scope of the current work.
This study strongly supports the hypothesis that hospitalists can increase the efficiency of care, even in systems already operating at high efficiency. Although baseline LOS and costs for SMHMO were among the lowest studied, significant improvements were seen after introduction of a hospitalist system. Moreover, overall ratings of hospitalization and ratings of key care processes improved after introduction of a hospitalist system, and continuity and quality of care did not suffer. However, we evaluated only relatively coarse quality measures, such as mortality. Additional studies are needed to measure the impact of hospitalists on other important parameters of quality, particularly evidence-based diagnostic testing and therapy, medical errors and adverse events, and objective measures of clinical care processes and outcomes. Studies of pediatric hospitalists should be performed in a variety of settings, and their relative influence on patients with varying medical conditions, socioeconomic backgrounds, and insurance plans measured. In teaching hospitals, it will be critical to assess the long-term impact of hospitalist systems on teaching, clinical research, and academic productivity and advancement.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Reprint requests to (C.P.L.) Children’s Hospital, Main 10E, Room 1032.2 300 Longwood Ave, Boston, MA 02115. E-mail: landrigan_c{at}hub.tch.harvard.edu
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  D. E. Wesson, D. Stephens, K. Lam, D. Parsons, L. Spence, and P. C. Parkin Trends in Pediatric and Adult Bicycling Deaths Before and After Passage of a Bicycle Helmet Law Pediatrics, September 1, 2008; 122(3): 605 - 610. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. N. Southern, M. A. Berger, E. Y. Bellin, S. M. Hailpern, and J. H. Arnsten Hospitalist Care and Length of Stay in Patients Requiring Complex Discharge Planning and Close Clinical Monitoring Arch Intern Med, September 24, 2007; 167(17): 1869 - 1874. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Srivastava, C. P. Landrigan, D. Ross-Degnan, S. B. Soumerai, C. J. Homer, D. A. Goldmann, and S. Muret-Wagstaff Impact of a Hospitalist System on Length of Stay and Cost for Children With Common Conditions Pediatrics, August 1, 2007; 120(2): 267 - 274. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. L. Freed, K. Brzoznowski, K. Neighbors, I. Lakhani, and American Board of Pediatrics, Research Advisory Co Characteristics of the Pediatric Hospitalist Workforce: Its Roles and Work Environment Pediatrics, July 1, 2007; 120(1): 33 - 39. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Boyd III, K. Samaddar, L. Parra-Roide, E. P. Allen, and B. White Comparison of Outcome Measures for a Traditional Pediatric Faculty Service and Nonfaculty Hospitalist Services in a Community Teaching Hospital Pediatrics, October 1, 2006; 118(4): 1327 - 1331. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. H. Conway, S. Edwards, E. R. Stucky, V. W. Chiang, M. C. Ottolini, and C. P. Landrigan Variations in Management of Common Inpatient Pediatric Illnesses: Hospitalists and Community Pediatricians Pediatrics, August 1, 2006; 118(2): 441 - 447. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. P. Landrigan, P. H. Conway, S. Edwards, and R. Srivastava Pediatric Hospitalists: A Systematic Review of the Literature Pediatrics, May 1, 2006; 117(5): 1736 - 1744. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. S. Lye, D. A. Rauch, M. C. Ottolini, C. P. Landrigan, V. W. Chiang, R. Srivastava, S. Muret-Wagstaff, and S. Ludwig Pediatric Hospitalists: Report of a Leadership Conference Pediatrics, April 1, 2006; 117(4): 1122 - 1130. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Coffman and T. G. Rundall The Impact of Hospitalists on the Cost and Quality of Inpatient Care in the United States: A Research Synthesis Med Care Res Rev, August 1, 2005; 62(4): 379 - 406. [Abstract] [PDF]
|
|
|
|
 |
|
 T. T.H. Nguyen Physicians' Attitudes Regarding Pediatric Hospitalist Systems AAP Grand Rounds, June 1, 2005; 13(6): 66 - 66. [Full Text] [PDF]
|
|
|
|
|
|
Section on Hospital Medicine Guiding Principles for Pediatric Hospitalist Programs Pediatrics, April 1, 2005; 115(4): 1101 - 1102. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Srivastava, C. Norlin, B. C. James, S. Muret-Wagstaff, P. C. Young, and A. Auerbach Community and Hospital-Based Physicians' Attitudes Regarding Pediatric Hospitalist Systems Pediatrics, January 1, 2005; 115(1): 34 - 38. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Dwight, C. MacArthur, J. N. Friedman, and P. C. Parkin Evaluation of a Staff-Only Hospitalist System in a Tertiary Care, Academic Children's Hospital Pediatrics, December 1, 2004; 114(6): 1545 - 1549. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Cash Pediatricians integral to coordinated care of hospitalized children: AAP policy stresses teamwork AAP News, March 1, 2003; 22(3): 118 - 120. [Full Text] [PDF]
|
|
|
|
 |
|
 Pediatric Hospitalists: Effect on Length of Stay and Cost Journal Watch (General), October 25, 2002; 2002(1025): 7 - 7. [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
