| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
ORIGINAL RESEARCH |
From the Section of Maternal-Fetal Medicine and Department of Obstetrics and Gynecology, Northwestern Memorial Hospital, Northwestern University Medical School, Chicago; and the Institute for Health Services Research and Policy Studies, Northwestern University, Chicago, Illinois.
Address reprint requests to: Susan E. Gerber, MD, MPH, Northwestern Memorial Hospital, 333 East Superior Street, Suite 410, Chicago, IL 60611; E-mail: s-gerber{at}northwestern.edu
 |
ABSTRACT |
|---|
TOP
 ABSTRACT MATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
METHODS: The proportions of low birth weight (LBW) and very low birth weight (VLBW) deliveries were tabulated for each hospital in Washington State for the years 1989, 1993 and 1996. Level of perinatal care, degree of health maintenance organization (HMO) penetration, and maternal demographic characteristics including age, race, smoking, and Medicaid status were derived from state and national databases. Multiple linear regression analysis was performed for each hospital level to evaluate the association between change in proportion of LBW and VLBW deliveries and change in HMO penetration per hospital between each of the 3 years.
RESULTS: From 1989 through 1993, the proportion of LBW deliveries significantly declined at level III hospitals and rose at level I and II hospitals. This trend reversed between 1993 and 1996. Very low birth weight deliveries demonstrated more limited and somewhat contrary results, significantly decreasing, then increasing in level I hospitals, and significantly increasing in level III hospitals from 1989 to 1993. After controlling for changes in maternal characteristics over time, changes in HMO penetration at the hospital level were not significantly associated with an increasing proportion of LBW or VLBW deliveries at nonlevel III hospitals. In some analyses, increasing HMO penetration actually was significantly associated with decreasing LBW and VLBW deliveries at nonlevel III hospitals.
CONCLUSION: Despite continued growth in HMOs throughout the state, the trend toward deregionalization in Washington State noted in the early 1990s has not continued. At the hospital level, the increasing presence of HMOs is not significantly associated with perinatal deregionalization.
The development of regionalized perinatal health services in the United States began in the 1970s in an effort to reduce neonatal mortality rates. Beginning in the 1980s, there has been a growing perception of a deregionalization of perinatal care.1–3 Powell et al3 demonstrated a significant change in site of delivery for low birth weight (LBW) neonates in Washington State in the late 1980s, with increasing numbers in nontertiary care (levels I–II) hospitals and decreasing numbers in tertiary care (level III) hospitals. Mehta et al2 similarly demonstrated a significant deregionalization trend in two regions in Ohio from 1990 to 1995.
The changing health care environment and the advent of managed care have often been cited as important causes of deregionalization. The role of managed care in the health care market has grown rapidly, with the national percentage of employees enrolled in managed care plans rising from 25% in 1987 to 75% in 1996.4 Most studies of maternal and neonatal outcomes do not demonstrate adverse effects associated with managed care.5,6 However, there are studies suggesting a negative effect of managed care on health outcomes in high-risk populations.4,7,8
The purpose of this study was to assess the extent and determinants of the deregionalization of perinatal health care. Washington State was chosen for analysis in light of previous literature demonstrating deregionalization in the early 1990s. In particular, the study focused on two questions: did a trend toward deregionalization persist from 1989 through 1996, and did an association exist between the increasing presence of managed care and the deregionalization of perinatal care?
|
MATERIALS AND METHODS |
|---|
|
TOP
ABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Delivery data were aggregated to the hospital level. Mean maternal age and the percentage of mothers who were smokers, black, or listed Medicaid as a source of prenatal care payment were calculated for use as risk adjusters for the hospital level analyses. Obstetric level of hospital was identified using data supplied by the Washington State Department of Health, Department of Vital Statistics, and confirmed by telephone calls to each hospital. Managed care penetration was defined by its most restrictive form, the health maintenance organization (HMO). Health maintenance organization penetration measured at the county level was obtained from the HMO Interstudy database.9 The HMO data were matched with the hospital data by county.
The primary study outcome was the extent of perinatal regionalization for high-risk neonates. For this study, perinatal deregionalization was defined as a shift in site of delivery of LBW or VLBW neonates from level III hospitals to level I and level II hospitals. Perinatal regionalization was measured on the statewide level (proportion of all LBW and VLBW deliveries occurring at each level of hospital) to identify trends in regionalization, and on the hospital level (proportion of deliveries at a given hospital that were LBW or VLBW).
Chi-square tests of proportions were conducted to determine whether significant changes in the extent of regionalization statewide occurred and to analyze the variation in HMO penetration by county. Linear regression was conducted with the hospital as the unit of analysis to identify factors that affected the proportion of deliveries at each hospital that were LBW for 1989, 1993, and 1996. Selected risk factors for LBW delivery (maternal age, black race, smoking, and Medicaid status)10 were used as independent variables to risk-adjust the hospitals for comparison, in order to allow for analysis of the effect of HMO penetration. Further linear regression was performed using the change in the proportion of LBW deliveries as the dependent variable, and the change in HMO penetration, and change in the other hospital demographic variables as the independent variables. The change regressions were estimated for level I and level II hospitals separately. Similar analyses were also conducted for the proportion of deliveries that were VLBW, using the log of the proportion.
Because clustering effects are introduced by the measurement of HMO penetration at the county level, the Huber correction was applied to all regressions to adjust the regression coefficient standard errors.11 This correction weights the standard errors by the extent of clustering in the HMO variable. The study was approved by the Northwestern University Institutional Review Board. All analyses were performed using SPSS 10.0 (SPSS Inc., Chicago, IL) and SAS 8 (SAS Institute Inc., Cary, NC) software.
|
RESULTS |
|---|
|
TOP
ABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Table 1
summarizes the percentage of LBW and VLBW babies born at each level of hospital during the 3 years of the study period. The percentage of LBW deliveries occurring at level III hospitals dropped significantly between 1989 and 1993, and increased between 1993 and 1996. The proportion of LBW deliveries at level I and level II hospitals increased and then decreased significantly during the same periods. Analysis of VLBW deliveries demonstrates a different trend, with VLBW births in level I hospitals decreasing between 1989 and 1993, then increasing between 1993 and 1996. A significant increase in VLBW deliveries at level III hospitals was noted between 1989 and 1993.
|
. Reported maternal smoking decreased from 31.7% in 1989 to 20.5% in 1996 (P < .001). Average maternal age increased from 26.4 to 27.1 years (P < .01). There was no significant change in the percentage of mothers who were black (from 4.0 to 4.5%). For all 3 years, the LBW proportion of the hospitals ranged from 0 to 0.23 [mean = 0.05, standard deviation (SD) = 0.03]. The VLBW proportion ranged from 0 to 0.08 (mean = 0.006, SD = 0.01).
|
summarizes the linear regression results (conducted with the hospital as the unit of analysis) estimating the impact of HMO penetration on the LBW proportion of a hospital for each year of the study. As expected, a lower proportion of LBW deliveries was significantly associated with a hospital being a level I or II hospital, and having a lower percentage of black mothers. However, HMO penetration was not a significant factor in determining this proportion for any of the 3 study years.
|
reports the results of regression analyses to determine the extent to which change in the LBW proportion at level I and level II hospitals was related to change in HMO penetration. Change regressions were not conducted for level III hospitals because of the small (n = 6) number of these hospitals. From 1993 to 1996, the decrease in the proportion of LBW deliveries at level I hospitals was significantly associated with an increase in HMO penetration at these hospitals. However, a significant association was not seen from 1989 through 1993, nor was it observed in analyses of the level II hospitals.
|
|
DISCUSSION |
|---|
|
TOP
ABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Washington State has a formal system of perinatal regionalization that has been in place for nearly 2 decades.16 In demonstrating a decline in perinatal regionalization in Washington State, Powell et al3 documented a potential cause for concern. For this reason, the present study was conducted with data from Washington State, to further extend their analysis and to investigate the potential causes of their findings. Washington State was also chosen as it has experienced a significant growth in managed care throughout the duration of the study period.17
Whereas perinatal deregionalization is widely assumed to be an ongoing trend, this study introduces new data from Washington State demonstrating that this assumption is not the case. Although we noted a significant decline in the percentage of LBW neonates delivered at level III hospitals through the early 1990s, the 1996 data indicate that the trend has reversed itself. Furthermore, this trend was not observed in the VLBW population, where the beneficial effects of regionalization are likely to be most important.
The mission of managed care is to contain costs without sacrificing patient care quality. Patient care, including maternity care, is more costly at tertiary care facilities. It is therefore understandable that in the interests of short-term cost-savings, managed care organizations might be disinclined to transfer patients from lower-cost to higher-cost facilities. However, in this study, no significant association was found between managed care penetration and perinatal deregionalization. In fact, although most analyses demonstrated no significant association, two actually demonstrated a positive significant association between managed care and regionalization. Health maintenance organization penetration was significantly associated with a decrease in the proportion of LBW deliveries at level I hospitals from 1993 to 1996, and of VLBW deliveries at level I hospitals form 1989 to 1993.
A number of potential reasons may be responsible for these findings. First, it is possible that managed care organizations encourage high-risk patients to deliver at tertiary care facilities. Recognition of decreased costs associated with decreased morbidity may provide the cost-savings incentive for managed care organizations to retain a system of regionalization. Patients may also be diverted to alternative level III hospitals that are in their managed care networks, but not in their traditional perinatal networks. Such an effect would not have been detected by this study. However, the ultimate goal of delivering at-risk patients at level III hospitals would still be met, and as such, this effect would be clinically irrelevant.
The influence of managed care may be felt at many levels—the hospital, the doctor, or the individual patient. For the purpose of this study, the hospital was chosen as the unit of analysis, because hospitals often have specific policies governing the transfer of patients, and are likely to be sensitive to increased pressures from managed care organizations. It is possible, however, that managed care may exert more of an influence on an individual patient level, and the effect would not be as apparent on a hospital level analysis.
There are a number of limitations to this study. The use of hospital-level analysis limits the power of this study, and 51 hospitals may have been insufficient to find significant differences in variables associated with shifts in regionalization. The use of county-level data may also introduce error, as counties may be poor proxies for hospital market areas. The demographic variables that were used to risk-adjust the hospitals were few in number, and did not always demonstrate significance in this model. The study was conducted in one state, with a distribution of HMO penetration and perinatal regional networks that may not be representative of other states. Finally, numerous other factors that may play a role in the extent of perinatal regionalization were not explored in this study.
Further studies are needed to understand the trends in regionalization in Washington State and throughout the country. At present, the managed care industry continues to evolve, and policies governing consumer health care continue to change. Ultimately, health outcomes such as maternal and neonatal mortality and morbidity rates should dictate perinatal health policies with an eye toward cost-effective health care.
|
Footnotes |
|---|
Received November 29, 2000. Received in revised form February 26, 2001. Accepted March 14, 2001.
|
REFERENCES |
|---|
|
TOP
ABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
2. Mehta S, Atherton HD, Schoettker PJ, Hornung RW, Perlstein PH, Kotagal UR. Differential markers for regionalization. J Perinatol 2000;20:366–72.[Medline]
3. Powell SL, Holt VL, Hickok DE, Easterling T, Connell FA. Recent changes in delivery site of low-birth-weight infants in Washington: Impact on birth weight-specific mortality. Am J Obstet Gynecol 1995;173:1585–92.[Medline]
4. Hellinger FJ. The effect of managed care on quality: A review of recent evidence. Arch Int Med 1998;158: 833–41.
5. Oleske DM, Linn ES, Nachman KL, Marder RJ, Sangl JA, Smith T. Effect of Medicaid managed care on pregnancy complications. Obstet Gynecol 2000;95:6–13.
6. Ray WA, Gigante J, Mitchel EF, Hickson GB. Perinatal outcomes following implementation of TennCare. JAMA 1998;279:314–6.
7. Bienstock JL, Blakemore KJ, Wang E, Presser D, Misra D, Pressman EK. Managed care does not lower costs but may result in poorer outcomes for patients with gestational diabetes. Am J Obstet Gynecol 1997;177:1035–7.[Medline]
8. Phibbs CS, Bronstein JM, Buxton E, Phibbs RH. The effects of patient volume and level of care at the hospital of birth on neonatal mortality. JAMA 1996;276:1054–9.[Abstract]
9. Wholey DR, Feldman R, Christianson JB. The effect of market structure on HMO premiums. J Health Econ 1995; 14:81–105.[Medline]
10. Creasy RK, Iams JD. Preterm labor and delivery. In: Creasy RK, Resnik R, eds. Maternal-fetal medicine. 4th ed. Philadelphia: W.B. Saunders, 1999:498–531.
11. Moulton BR. An illustration of a pitfall in estimating the effects of aggregate variables on micro units. Rev Econ Stat 1990;72:334–8.
12. Paneth N, Kiely JL, Wallenstein S, Susser M. The choice of place of delivery: Effect of hospital level on mortality in all singleton births in New York City. Am J Dis Child 1987; 141:60–4.[Abstract]
13. Gortmaker S, Sobol A, Clark C, Walker DK, Geronimus A. The survival of very low-birth weight infants by level of hospital birth: A population study of perinatal systems in four states. Am J Obstet Gynecol 1985;152:517–24.[Medline]
14. McCormick MC, Shapiro S, Starfield BH. The regionalization of perinatal services: Summary of the evaluation of a national demonstration program. JAMA 1985;253: 799–804.[Abstract]
15. Yeast JD, Poskin M, Stockbauer JW, Shaffer S. Changing patterns in regionalization of perinatal care and the impact on neonatal mortality. Am J Obstet Gynecol 1998;178: 131–5.[Medline]
16. Rosenblatt RA, Macfarlane A, Dawson AJ, Cartlidge PHT, Larson EH, Hart LG. The regionalization of perinatal care in Wales and Washington State. Am J Public Health 1996;86:1011–5.
17. Simon CJ, Dranove D, White WD. The impact of managed care on the physician marketplace. Public Health Rep 1997;112:222–30.[Medline]
This article has been cited by other articles:
|
|
 |
|
  Z. Kabir, G. N. Connolly, L. Clancy, B. B. Cohen, and H. K. Koh Declining maternal smoking prevalence did not change low birthweight prevalence in Massachusetts from 1989 to 2004 Eur J Public Health, November 5, 2008; (2008) ckn106v1. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Dobrez, S. Gerber, and P. Budetti Trends in Perinatal Regionalization and the Role of Managed Care. Obstet. Gynecol., October 1, 2006; 108(4): 839 - 845. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
