 |
|
Early Postpartum Discharges
Impact on Distress and Outpatient Problems
David A. Lane, MD;
Lynda S. Kauls, MD;
Jeannette R. Ickovics, PhD;
Frederick Naftolin, MD, PhD;
Alvan R. Feinstein, MD
Arch Fam Med. 1999;8:237-242.
ABSTRACT
| |
Objective To determine the impact of shortened postpartum hospital stays on common clinical phenomena in a sociodemographically diverse, unselected group of general maternity patients.
Design Observational cohort study in which the preapproved hospital stay duration of either 1 or 2 nights was set by third-party payers before each mother's admission.
Setting Yale–New Haven Hospital, New Haven, Conn, from June 19 through August 10, 1995.
Patients Two hundred forty-four volunteers from among 400 eligible deliveries.
Main Outcome Measures Readmission within 1 month of hospital discharge, report of outpatient morbidity and use of outpatient health services within 1 week of discharge, status of breast-feeding during the first postdischarge week, and patient satisfaction.
Results At discharge from the hospital, the hospital stay was regarded as "too short" by 80 (47%) of 171 mothers and 19 (26%) of 73 mothers in the 1- and 2-night groups, respectively (P=.002). Although readmission rates were similar (5% vs 3%, P=.48), the 1-night group reported significantly more morbidity in the newborns (31% vs 16%, P=.03) and averaged more pediatric visits (96 vs 54 per 100 newborns, P=.002). Mothers in the 1-night group also reported more fatigue (49% vs 29%, P=.001) and more worries about their newborns' health (24% vs 11%, P=.02). They were less likely to start breast-feeding (64% vs 77%, P=.06), and, if they started, were somewhat more likely to stop prematurely (14% vs 8%, P=.43). A series of disturbing events was reported only in the 1-night group.
Conclusions In a relatively unselected group, mothers who stayed 1 night after routine vaginal delivery reported more distress and more pediatric problems and had greater use of outpatient health services than mothers who stayed 2 nights.
INTRODUCTION
THE AVERAGE length of stay (LOS) in most US hospitals after a routine vaginal delivery has been progressively reduced to 1 night or less.1-3 Although the shorter stays save costs, their safety has been debated. For example, several studies4-7 and 3 randomized trials8-10 have shown no effect on readmissions, but a recent population-based case-control study11 reported that newborns discharged from the hospital less than 30 hours after birth were more likely to be readmitted than newborns who stayed longer. According to some contentions, however, readmissions and similar hard data do not reflect the clinical and human impact on patients who may be sent home too soon.12-17
The supporting evidence for and against the short-stay policy is influenced by several important factors. First, both the prospective trials and retrospective studies have limited their results to a narrow scope of uncommon outcome events (ie, readmissions). Second, the prospective investigations selected prescreened, low-risk volunteer subjects who could refuse their randomly assigned discharge time, and excluded many mothers and newborns with relatively common peripartum difficulties. Finally, the impact of postdischarge events, such as maternal stress or fatigue, early scheduled outpatient follow-up, or regular home visits by a nurse, was not addressed.
In the current research, we used a prospective, observational study to examine pertinent clinical phenomena, including readmissions and other postdischarge events, for a relatively unselected general maternity population. Random assignment to a 1- or 2-night postpartum stay was approximated by the preapproved LOS set by third-party payers before each mother's admission—an assignment unaffected by patients' predelivery sociodemographic attributes or by their postdelivery clinical course.
SUBJECTS AND METHODS
STUDY POPULATION
The study was done at Yale–New Haven Hospital, New Haven, Conn, the principal teaching hospital for Yale University School of Medicine. The patients represent all sociodemographic groups and account for almost 4000 vaginal deliveries annually, nearly three fourths of all routine deliveries in greater New Haven.
Mothers eligible for the study must have had vaginal delivery of a singleton term infant, routine postpartum care, and a physician's order for discharge to home. Patients were excluded for unwillingness to participate, inability to communicate in English, clinical complications that added at least 1 night to the preauthorized duration of 1 or 2 nights (stays of 3 or more nights were immediately excluded), or complications that delayed the newborns' discharge from the hospital more than 24 hours beyond the mother's actual departure.
PROCEDURES
Eligible patients were enrolled in this human investigations committee–approved study from June 19 through August 10, 1995. After reviewing discharge orders each day at the hospital's 2 postpartum units, trained interviewers invited eligible patients, for whom discharge from the hospital was imminent, to participate. When the number of discharges exceeded the team's usual capacity of about 6 interviews per day, invitations and interviews were randomly chosen by patients' room numbers. Despite efforts to interview every eligible mother regardless of departure time, the high volume of deliveries inevitably allowed some patients to leave the hospital before they could be invited to participate.
Regardless of their actual LOS, invited patients were informed only that the study was intended to assess what happens to mothers and newborns after discharge from the hospital, and were advised that participation or nonparticipation would have no effect on the care they or their newborns received. Local physicians were unaware of the study's hypotheses and selection criteria or whether their patients were participating.
The participating mothers received face-to-face interviews just before leaving the hospital, with follow-up telephone interviews 1 week and 1 month later. Each interview adhered to a written script of specific questions that allowed open-ended answers and lasted about 20 to 25 minutes. Every effort was made to balance the interview by asking any polarized open-ended questions in both directions (eg, "What did you like most about . . . ?" and "What did you like least about . . . ?"). In concert with the written script, the 7 interviewers (2 full time and 5 part time) encouraged the mothers to speak openly and freely when responding to the questions. In addition, whenever possible, all 3 interviews with each patient were conducted by the same interviewer, who recorded detailed notes of the patient's comments, including specific phrases that could be grouped into similar themes during data analysis.
The first interview (at discharge from the hospital) sought to determine baseline sociodemographic and clinical characteristics as well as maternal fatigue, support, and satisfaction. Participants were asked what they liked and disliked about their postpartum experiences and about personal reactions to their hospital stay. The 2 follow-up interviews emphasized intervening maternal and neonatal health-related events. For example, mothers were asked to describe problems or difficulties they or their newborns experienced after discharge from the hospital, and how they had managed them. If a problem or difficulty involved a call or visit to a health care provider, further questions were asked to elicit any objective findings that had been noted, as well as the provider's assessment and treatment plan. We counted a cited problem as a reported morbidity if it occurred within 1 week of discharge from the hospital and evoked either specific outpatient treatment or at least 2 visits and/or calls to a physician or nurse-practitioner. In addition to tracking morbidity, this process allowed us to tally all health care services mothers reported using during the first postdischarge week and reasons for using the service, including home visits, telephone calls, and visits to physicians and other providers (ie, nurse-practitioners, midwives, and lactation consultants). Continuation of breast-feeding (if pertinent) was also determined.
DIVISION INTO COMPARISON GROUPS
Each patient anticipating a vaginal delivery entered the hospital with a preauthorized LOS of either 1 or 2 nights, according to the policies of the third-party payers. After all the initial and follow-up interviews had been completed, we obtained a list from hospital administrators showing the particular durations specified by the various third-party payers and verified that each mother's actual LOS was consistent with the preauthorized LOS. We defined a full hospital night as 1 overnight period in the hospital beginning before 9 PM. Participating mothers, all of whom stayed 1 or 2 hospital nights, were stratified according to their LOS into the 2 groups. Thus, although the actual duration of the mothers' postpartum stay was learned during the course of open-ended questioning, neither the investigators nor the interviewers were aware of the comparison group for a particular mother-newborn pair until after all the data were collected.
Because mothers with stay-lengthening complications were excluded, the preadmission assignments were carried out in all but 30 members of the studied group. The exceptions occurred in 2 mothers authorized for 2 nights who insisted on leaving after 1 night, and in 28 Medicaid patients who were discharged from the hospital after 1 night because this had become the hospital routine for such patients (many physicians were unaware of specific Medicaid entitlements). These 30 patients were analyzed according to their actual rather than authorized duration. Their exclusion from the subsequent statistical analyses either had no effect or increased existing differences between the compared groups.
STATISTICAL ANALYSES
We used SAS statistical software (SAS Inc, Cary, NC) to analyze continuous and ordinal data with 2-tailed t tests and Wilcoxon rank sum tests, and relative frequencies of occurrence for categorical data with  2 or Fisher exact tests. All pertinent 2-tailed P values are cited.
RESULTS
STUDY POPULATION
Among 564 vaginal deliveries at Yale–New Haven Hospital during the enrollment period, 406 had no stay-lengthening complications and met the eligibility criteria, except for 6 mothers who did not speak English. Of the remaining 400 mothers, 105 departed before invitation or interview, and 51 declined to participate. The 244 mother-newborn pairs who received the first interview became the study cohort, with 171 mothers (70%) in the 1-night group and 73 (30%) in the 2-night group. Table 1 shows that the 2 groups had similar baseline sociodemographic and clinical characteristics, including anticipation of adequate help at home. Although the apparent difference in the proportion of primiparas was not statistically significant (P=.10), all subsequent comparisons were checked separately and showed no pertinent differences between the primiparous and multiparous subgroups.
|
|
|
|
Table 1. Sociodemographic and Clinical Features of 244 Mother-Newborn Pairs With Uncomplicated Vaginal Delivery at Yale–New Haven Hospital
|
|
|
NONPARTICIPANT AND FOLLOW-UP LOSSES
Of the 295 invited mothers, 51 (17%) declined to participate. The cited reasons were unwillingness to be a study subject in 27; reluctance to divert time from other activities, such as attending classes on infant care and feeding, napping, and visiting with family and friends, in 21; and unavailability for follow-up interviews in 3. Of 156 eligible nonparticipants, 101 (65%) stayed 1 postpartum night and 55 (35%) stayed 2 nights, so that the distributions of nonparticipants and participants were similar in the 2 groups. The nonparticipants did not differ substantially from the participants in number of nights stayed, age, parity, AM vs PM departure, or type of health plan.
The first follow-up interview was completed by 230 (94%) of the 244 cohort mothers and the final interview by 211 (86%). Rates of noncompletion at 1 month (14%) were identical for the 2 compared groups.
READMISSIONS AND REPORTS OF OUTPATIENT MORBIDITY
Table 2 shows that during the 1-month follow-up period, 8 newborns were readmitted from the 1-night group and 2 from the 2-night group. The readmissions were for suspected sepsis in 6 cases, for jaundice in 3, and for dehydration in 1. No mothers from either group were readmitted.
|
|
|
|
Table 2. Hospital Admissions Within 1 Month and Reports of Outpatient Morbidity Within 1 Week of Postpartum Discharge From Yale–New Haven Hospital*
|
|
|
As also shown in Table 2, however, morbidity was reported during the first postdischarge week for 50 newborns (31%) in the 1-night group and for 11 (16%) in the 2-night group (P=.03). Jaundice, nutrition problems (weight loss or dehydration), and mucocutaneous lesions—noted in 21, 13, and 12 newborns, respectively—were the most frequently reported categories of morbidity and were proportionately more common in the 1-night group. Three diagnostic categories were reported only among newborns in the 1-night group, including 6 problems with the umbilical cord (in 4, the clamp was inadvertently not removed), 6 cases of gastrointestinal or respiratory difficulty, and 2 problems with circumcision.
Morbidity was reported for 21 mothers (13%) in the 1-night group and 6 (9%) in the 2-night group. The leading causes were impediments to lactation (mastitis or clogged milk duct), vaginal infections or bleeding, and postdelivery pain, cited by 6, 5, and 4 mothers, respectively.
USE OF OUTPATIENT HEALTH SERVICES
As shown in Table 3, the use of medical services (office visits to physicians and nurse-practitioners) was significantly greater in the 1-night group than in the 2-night group (96 vs 54 per 100 newborns, P=.002). Even though some pediatricians routinely saw early-discharge newborns in their offices a day or so after discharge, the difference persisted for both scheduled and unscheduled visits. The rate of telephone calls for pediatric advice was 81 per 100 newborns for the 1-night group and 79 for the 2-night group. During the same period, outpatient health services were sought less often for mothers than for their newborns. The mothers in the 2-night group, however, made substantially more telephone calls about themselves than mothers in the 1-night group. The difference may have arisen because the 1-night group had more scheduled opportunities for communication with the health care system. Patients in the 1-night group were offered and received significantly more home visits by a nurse. Mothers said that the offer of a home visit, the number of visits offered, and the skill level of the visiting provider (ie, nurse, aide, nanny) depended more on the specific features of a health plan than on their individual needs. Although most mothers found the home visits reassuring and helpful, we found no reduction in readmissions, in reported outpatient morbidity, or in use of other outpatient medical services among those who received home visits by a nurse.
|
|
|
|
Table 3. Use of Outpatient Health Services Within 1 Week of Discharge*
|
|
|
Table 4 shows the individual frequency of home, office, and nursing visits during the first week after discharge from the hospital. Mothers and newborns in the 1-night group were not only more likely to visit their physicians during the first postdischarge week, but they also often made more than 1 visit. For example, 43 newborns (26%) in the 1-night group made 2 or more office visits, compared with 7 (10%) in the 2-night group. The tendency for multiple visits to physicians during the first week did not carry over to home nursing visits. Some mothers and newborns in both groups had no follow-up contact of any kind (ie, calls, office visits, and home visits). In the 1-night group, 23 mother-newborn dyads (14%) had no medical follow-up, compared with 11 (16%) in the 2-night group (P=.68).
|
|
|
|
Table 4. Home, Office, and Nursing Visits Within 1 Week of Discharge*
|
|
|
BREAST-FEEDING STATUS
Breast-feeding was started in the hospital by 63% (108/171) of mothers in the 1-night group and by 67% (49/73) in the 2-night group. Although all breast-feeding mothers had planned to continue for at least 2 weeks, among those who completed follow-up interviews within the first 7 days after discharge from the hospital, 14% (15/108) in the 1-night group and 8% (4/49) in the 2-night group had stopped. Receipt of a home visit had no discernible impact on continuation or cessation of breast-feeding.
PATIENT SATISFACTION
Eighty-one mothers (47%) in the 1-night group said that their stay was "too short," compared with 19 (26%) in the 2-night group (P=.002). The reasons for satisfaction (or dissatisfaction) could be grouped, as shown in Table 5, into 3 main categories we have called perimorbidity. The comment cited most often—not having enough rest—was mentioned by 84 mothers (49%) in the 1-night group and by 21 (29%) in the 2-night group (P=.001). The next most frequently reported concern—expressed by 41 mothers (24%) in the 1-night group and by 8 (11%) in the 2-night group (P=.02)—was the worry that physicians had not had enough time to observe the newborn for health problems. The third most common concern was anxiety that more supervised instructions from physicians and/or nurses were needed for the mothers to properly care for their newborns. In the 1-night group, 31 (18%) had anxiety about their competence, compared with 10 (14%) in the 2-night group.
|
|
|
|
Table 5. Mothers' Main Concerns Just Before Departure From Hospital
|
|
|
NOTABLE EVENTS
In response to the open-ended questions, no mothers from either group reported any particularly happy experiences related to their LOS, and no mothers from the 2-night group reported any distressful episodes. Several mothers in the 1-night group, however, spontaneously recounted upsetting events that they attributed to their LOS.
For example, at least 4 mothers were sent home weeping. In 3 instances they had perineal pain and/or swelling from delivery, including 1 fourth-degree laceration (extending to the rectum). These mothers had difficulty walking and said they did not feel physically ready to take care of their newborns at home. Another tearful woman, unable to void urine, had to be catheterized by her husband at home because she was too distressed to do it herself as instructed. The distressed women said their physicians' requests for 1 more hospital night had been denied as being "comfort care."
Although discharged from the hospital during morning rounds, many mothers lingered in their hospital rooms until late at night. Among the late-staying mothers, at least 10 said they wanted to give the physicians more time to observe the newborns for health problems, and 7 said they needed more supervised teaching on taking care of their newborns, particularly about breast-feeding.
Eight families in the 1-night group had "staggered" discharges from the hospital, which sent the mother home in the evening while the newborn was hospitalized overnight for more observation. Two of these families used their own funds to pay for a second night's stay for the mother, and several breast-feeding mothers stayed in the hospital lobby for extended periods between feedings. Another mother slept on a cot in the nursery. All of the staggered-discharge newborns were released from the nursery the next morning, often less than 12 hours after the mother's discharge from the hospital.
COMMENT
This study provides outcome data for a broad spectrum of patients subjected to the clinical and human impact of so-called "drive-through deliveries." To minimize problems of recruitment and selection, we relied on an observational study in which the counterpart of randomization was the reasonably unbiased predelivery assignment of a preauthorized postpartum LOS. In a pilot study, only 1 of the first 56 participants said that her choice of a health plan was affected by the length of inpatient postpartum coverage.
Although relatively small, the cohort resembles the general obstetrical population at many institutions, is, to our knowledge, larger than the cohort of any published, current, prospective study of early postpartum discharges from the hospital, and is more than twice the size of the groups in the 3 published randomized trials. Nevertheless, despite our efforts to include all mothers who had "routine" postpartum care after "uncomplicated" vaginal deliveries, the results may be skewed by the absence of the 51 mothers who declined to participate. Because of the nonparticipation, we have no information about the pertinent outcomes for these mothers. During the brief interview, however, some of them expressed dissatisfaction about their length of stay. Exclusion of the 6 women who did not speak English probably did not affect the results, although we do not know the impact of the cultural differences and expectations.
Although the study was conducted at a single medical center whose practice patterns may not be mirrored at all institutions with obstetrical services, the current research seems pertinent for everyday clinical practice. The observed postpartum and neonatal practice patterns seem rather typical, and the maternity patients came from many walks of life and had an array of different health plans. We included many mothers whose intrapartum and postpartum events would have excluded them from previous prospective studies.
We found that rates of readmission were not substantially increased in patients with shorter LOS, but the 1-night group had significantly more reported morbidity, perimorbidity, and outpatient visits to pediatricians than the 2-night group. Members of the 1-night group also described many distressing notable events that were not reported by the other mothers.
Although the drive-through delivery policy may be ending as legislative action is taken to lengthen the shortened stays,17-22 several useful lessons can be learned from the controversy. The outcomes from randomized trials and population-based studies can help establish general policy guidelines for a particular population, but may not be suitable for clinical decisions involving individual patients. In addition, although readmissions are important outcome events, they may not reflect clinical and personal effects that are meaningful to many patients (or their physicians). Finally, significant attributes of care and satisfaction are often best discerned by directly asking patients what they like and dislike about their care.
CONCLUSIONS
This study suggests that the ideal length of a postpartum stay may vary from one woman to the next. Some mothers are ready, willing, and able to go home after 1 night, whereas others are not. Therefore, a reasonable approach might be to keep most mothers and newborns hospitalized 2 nights, but to allow shorter stays when requested by patients and approved by their physicians.
AUTHOR INFORMATION
Accepted for publication May 28, 1998.
At the time of this research, Dr Lane, a family physician in the US Navy, was assigned as a fellow in the Yale University Clinical Scholars Program, funded by the Robert Wood Johnson Foundation, Princeton, NJ. Dr Lane has since been reassigned to the Primary Care Group, Naval Medical Center, San Diego, Calif.
The views expressed in this article are the authors' and do not reflect the official policy of the Department of the Navy, Department of Defense, or the US Government.
We thank Jennifer Bell and Yale medical students Karen Earle, A. J. Babineau, and Michelle Barton for data collection; Carolyn K. Wells for data analysis; Donna Cavaliere for manuscript preparation; and Vicki Lane for data collection and manuscript preparation.
Corresponding author: Alvan R. Feinstein, MD, Room I 456 SHM, Yale University School of Medicine, New Haven, CT 06520-8025.
From the Departments of Medicine (Drs Lane, Ickovics, and Feinstein) and Obstetrics and Gynecology (Dr Naftolin), Yale University School of Medicine, New Haven, Conn. When this research was done, Dr Kauls was a third-year medical student at Yale University School of Medicine.
REFERENCES
| |
1. Epidemiology Office of the Centers for Disease Control and Prevention. Trends in length of stay for hospital deliveries—United States, 1970-1992. MMWR Morb Mortal Wkly Rep. 1995;44:335-336.
PUBMED
2. Braveman P, Egerter S, Pearl M, Marchi K, Miller C. Early discharge of newborns and mothers: a critical review of the literature. Pediatrics. 1995;96:716-726.
FREE FULL TEXT
3. Length of Stay by DRG and Payment Source, United States, 1994. Baltimore, Md: Health Care Investment Analysts Inc; 1994:372-373.
4. Scupholme A. Postpartum early discharge: an inner city experience. J Nurs Midwifery. 1981;26:19-22.
5. Berryman GK. Early discharge of mothers and infants following vaginal childbirth. Mil Med. 1991;156:583-584.
ISI
| PUBMED
6. Pittard WB, Geddes KM. Newborn hospitalization: a closer look. J Pediatr. 1988;112:257-262.
FULL TEXT
|
ISI
| PUBMED
7. Lee KS, Perlman M, Ballantyne M, Elliot I, To T. Association between duration of neonatal hospital stay and readmission rate. J Pediatr. 1995;127:758-766.
FULL TEXT
|
ISI
| PUBMED
8. Waldenström U, Sundelin C, Lindmark G. Early and late discharge after hospital birth: health of the mother and infant in the postpartum period. Ups J Med Sci. 1987;92:301-314.
ISI
| PUBMED
9. Yanover M, Jones D, Miller M. Perinatal care of low-risk mothers and infants: early discharge with home care. N Engl J Med. 1976;294:702-705.
ABSTRACT
10. Carty E, Brady C. A randomized, controlled evaluation of early postpartum hospital discharge. Birth. 1990;17:199-204.
ISI
| PUBMED
11. Liu LM, Clemens CJ, Shay DK, Davis RL, Novack AH. The safety of newborn early discharge programs: the Washington State experience. JAMA. 1997;278:293-298.
FREE FULL TEXT
12. Annas G. Women and children first. N Engl J Med. 1995;333:1647-1651.
FREE FULL TEXT
13. Parisi VM, Meyer BA. To stay or not to stay? that is the question. N Engl J Med. 1995;333:1635-1637.
FREE FULL TEXT
14. Council on Scientific Affairs of the American Medical Association. Impact of 24-Hour Postpartum Stay on Infant and Maternal Health: Report 5 of the Council on Scientific Affairs (A-95). Chicago, Ill: American Medical Association; 1995. Available in summary form at: http://www.ama-assn.org/med sci/csa/1995/rpt4a95.htm. Accessed March 11, 1999.
15. Miller J. Mother and newborn: how long in the hospital? New York Times. August 20, 1995;sect 13:1.
16. Nordheimer J. Broad coalition fights insurer limits on hospital maternity stays. New York Times. June 4, 1995:B1.
17. Associated Press. Longer hospital stays for childbirth are needed, pediatricians say. New York Times. October 11, 1995:A17.
18. Newborns' and Mothers' Health Protection Act of 1995: Hearing Before the Committee on Labor and Human Resources, United States Senate, One Hundred Fourth Congress, First Session on S. 969, 12 Sept, 1995. Washington, DC: Government Printing Office; 1995.
19. Nordheimer J. New mothers gain 2nd day of care. New York Times. June 29, 1995:B1.
20. Powell A. Docs critical of insurance rules. New Haven Register. January 17, 1996:A1.
21. Hernandez R. Assembly passes a bill regulating childbirth stays. New York Times. January 17,1996:A1.
22. Page L. Clamping down on managed care. American Medical News. August 11, 1997:9, 14-15.
THIS ARTICLE HAS BEEN CITED BY OTHER ARTICLES
Newborn Length of Stay, Health Care Utilization, and the Effect of Minnesota Legislation
Madlon-Kay et al.
Arch Pediatr Adolesc Med 2003;157:579-583.
ABSTRACT
| FULL TEXT
Effects on Breastfeeding of Changes in Maternity Length-of-Stay Policy in a Large Health Maintenance Organization
Madden et al.
Pediatrics 2003;111:519-524.
ABSTRACT
| FULL TEXT
Newborn Early Discharge Revisited: Are California Newborns Receiving Recommended Postnatal Services?
Galbraith et al.
Pediatrics 2003;111:364-371.
ABSTRACT
| FULL TEXT
Quality of Life: Erosions and Opportunities Under Managed Care
Morreim
J Law Med Ethics 2000;28:144-158.
|
