Early hospital discharge and early puerperal complications
Egreso temprano postparto y complicaciones en el puerperio mediato
Dolores Ramírez-Villalobos, MScI; Adolfo Hernández-Garduño, DScII; Aarón Salinas, MScIII; Dolores González, MScI; Dilys Walker, MDI; Guadalupe Rojo-Herrera, MDII; Bernardo Hernández-Prado, DScI
ICentro de Investigación en Salud Poblacional. Instituto Nacional de Salud Pública. Cuernavaca, Morelos, México
IIHospital General de México, Secretaría de Salud, México, DF
IIICentro de Investigación en Evaluación y Encuestas. Instituto Nacional de Salud Pública. Cuernavaca, Morelos, México
OBJECTIVE:To evaluate the association between time of postpartum discharge and symptoms indicative of complications during the first postpartum week.
MATERIALS AND METHODS: Women with vaginal delivery at a Mexico City public hospital, without complications before the hospital discharge, were interviewed seven days after delivery. Time of postpartum discharge was classified as early (<24 hours) or late (>25 hours). The dependent variable was defined as the occurrence and severity of puerperal complication symptoms.
RESULTS:Out of 303 women, 208 (68%) were discharged early. However, women with early discharge and satisfactory prenatal care had lower odds of presenting symptoms in early puerperium than women without early discharge and inadequate prenatal care (OR 0.36; 95% confidence intervals = 0.17-0.76).
CONCLUSIONS:There was no association between early discharge and symptoms of complications during the first postpartum week; the odds of complications were lower for mothers with early discharge and satisfactory prenatal care.
Key words: patient discharge; postpartum period; postnatal care; Mexico
OBJETIVO:Evaluar la asociación entre el tiempo de egreso posparto y las posibles complicaciones en el puerperio mediato.
MATERIAL Y MÉTODOS:Mujeres con parto vaginal atendidas en un hospital público de la Ciudad de México, sin complicaciones antes del egreso hospitalario, fueron entrevistadas a los siete días de egreso. La variable dependiente fue la ocurrencia y severidad de complicaciones. Se calcularon media y desviación estándar para las variables continuas, y proporciones para las categóricas. Las variables relacionadas con egreso temprano en el análisis bivariado (con p<0.15) fueron incluidas en un modelo de regresión logística.
RESULTADOS:Se analizó información de 303 partos, de los cuales 208 (68%) tuvieron egreso temprano posparto. Las mujeres que fueron egresadas en forma temprana con un control prenatal adecuado reportaron menos síntomas de complicaciones en el puerperio mediato (RM= 0.36; IC 95% = 0.17-0.76).
CONCLUSIONES:Aunque no se encontró asociación entre el egreso temprano y los síntomas de complicaciones durante la primera semana del posparto, el riesgo de complicaciones fue menor en mujeres con egreso temprano y con cuidado prenatal adecuado, comparadas con las mujeres que presentaron egreso tardío sin control prenatal.
Palabras clave: alta del paciente; periodo de posparto; atención posnatal; México
In recent years, there has been growing interest to determine the ideal time for postpartum discharge for optimal maternal and child outcomes.1-9 Hospital length of stay after childbirth has decreased progressively during the past 60 years.10-15 In the early 1980s in Mexico, the Mexican Social Security Institute (Instituto Mexicano del Seguro Social - IMSS) developed the program Atención de Parto de Bajo Riesgo (Care for Low-Risk Delivery),16, 17 which resulted in a six-hour reduction in postpartum hospital stays.
For women who have uncomplicated vaginal deliveries, the American College of Obstetrics and Gynecology (ACOG) defines early discharge (ED) as a hospital stay lasting 48 hours or less, and considers a stay of 24 hours or less very early discharge (VED).18-21 The hypothesis of this study is that shorter postpartum stays are associated with poor health outcomes because of the decreased probability of detecting postpartum complications, as has been found in studies with other populations.22
Little information is available in Mexico to assess the potentially negative effects of early hospital discharge on maternal health during early puerperium, defined as the period between 24 hours to 7 days postpartum.23 It is important to assess whether mothers who are discharged early after vaginal delivery are at risk to develop complications, as well as what type of complications may occur. The aim of this study was to evaluate the association between time of postpartum discharge and reported symptoms indicative of complications during early puerperium.
Material and Methods
The study population consisted of women who received obstetric care after normal vaginal delivery at the Gynecology and Obstetrics Department of the General Hospital of Mexico (HGM), Mexico City, Ministry of Health (Secretaría de Salud, SSA) between April and December 2003.
The inclusion criteria for the study were: a) vaginal delivery of a live singleton term infant (gestational age 37 to 41 weeks); b) uncomplicated pregnancy without concomitant diseases such as diabetes, hypertension, preeclampsia, cardiopathy, epilepsy, or evident infections; c) routine postpartum care, and d) residence in Mexico City. The exclusion criteria were refusal to participate, checking out of the hospital separately from the child and residing outside the city. The withdrawal criteria were refusal to continue participating and failure to locate the patient after three attempts.
Six trained interviewers evaluated medical records to select subjects who fulfilled the inclusion criteria and then invited eligible mothers to participate in the study. After signed informed consent, selected mothers participated in a face-to-face interview prior to leaving the hospital to collect the following baseline data: a) sociodemographic characteristics; b) gynecologic and obstetric history; c) prenatal care assessed according to the Official Mexican Norm (NOM-007-SSA2-1993),24 d) delivery events, including vaginal lacerations; e) clinical characteristics of the immediate puerperium, (considered as the 24-hour period following delivery);24 and f) physician's discharge orders. A chart review was performed for all cases to corroborate questionnaire data and obtain clinical information. Upon discharge, mother-child pairs were invited for a medical visit seven days after delivery to assess newborn health status. At this visit, mothers underwent another face-to-face interview to obtain information related to maternal and infant postpartum health.
Women reporting serious complication symptoms were referred to the hospital's Gynecology and Obstetrics Service for clinical evaluation. In the event that the mother failed to attend the 7-day follow-up appointment, a trained interviewer visited her at home to complete the interview. The study was approved by the Ethics, Biosafety, and Research Committees of the Mexican National Institute of Public Health and of the General Hospital of Mexico.
The study outcome variable was the presence of self-reported symptoms in early puerperium. This variable was measured using symptoms reported by the mother during an interview conducted seven days after hospital discharge. Symptoms were categorized as suggestive of: a) urinary tract infection (dysuria, frequent urination, bladder tenesmus); b) episiotomy complications (local pain or discomfort, bleeding, separation of sutures, c) episiotomy infection (purulent discharge, pain, warmth and redness in the area); d) endometritis: (uterine pain, foul smelling lochia, and fever or shivering); f) mastitis and/or mammary abscess (pain, heat, and redness or cracking of nipples); and g) other reported symptoms or hospital readmission. Subjects were assigned one of two categories: a) absence of symptoms or b) presence of any symptom.
The exposure variable was the time of postpartum hospital discharge, measured as the time elapsed from delivery to hospital discharge (according to hospital records). For this study, early postpartum discharge (ED) was defined as 24 hours or less, whereas late discharge (LD) was defined as later than 25 hours. Potential confounders included sociodemographic or obstetric variables, perinatal and delivery events, and early postpartum complications.
It is important to clarify that in the facility where the study was conducted, physicians' discharge orders are given by the responsible obstetrician during clinical rounds that occur each day in the morning and afternoon. Depending on the clinical status of the patient the rounding physician will give the discharge order. These orders depend on non-clinical (bill must be paid prior to discharge and dedicated blood donation is sometimes required) as well as clinical (no apparent complication) indications for discharge. In some cases of non-clinical discharge, some women in the late discharge group were actually candidates for early discharge but were kept for non-clinical reasons.
Data are presented as mean and standard deviation for continuous variables and proportions for categorical variables. Bivariate associations between the outcome variable and each of the covariates were assessed to obtain odds ratios and 95% confidence intervals. The variables related to early discharge in the bivariate analysis at p<0.15 were included in a logistic regression model.
An interaction term was added to assess the potential modifying effect of ED with satisfactory prenatal care on the presence of complications in early puerperium. All women with any complication identified during the hospital stay were excluded from the analysis. There were four women whose hospital stays were longer than 72 hours (77, 79, 87 and 99 hours). The analysis was conducted both including and excluding those observations, and no differences were found. The results that we present in this paper include these four women.
Finally, regression diagnostics were obtained for the logistic model.25 The statistical analysis was performed using Stata Version 9.0.*
Of the 5 326 women who delivered at the HGM between April 11 and December 15, 2003, a total of 2 710 (50.8%) had normal vaginal deliveries; 829 of those were eliminated due to premature birth, low birth weight (d"2 600 g or d"36 weeks gestational age), stillbirths, or twins. Of the remaining 1 881 vaginal deliveries, 1 216 (64.6%) fulfilled all inclusion criteria. Of these women, 323 (26.6%) did not participate because they resided outside the metropolitan area or simply because they did not wish to participate in the study. No differences were found between women who agreed to participate and those who did not with respect to number of live born children, newborn weight, maternal age or time of discharge. From the 893 remaining women, 497 were excluded from the analysis because they had complications before the hospital discharge. Out of the 396 women without complications prior to hospital discharge, only 303 were included in the analysis because 93 did not have complete information on the variables under study (Figure 1). We did not find significant differences between women with complete and incomplete information regarding their age, length of hospital stay and number of live-born children.
During follow-up, 63 women (15.9%) were lost due to change of residence, failure to locate the place of residence after three attempts, or incorrect address. No significant differences were found between women who completed the study and those who were lost to follow-up, regarding the length of hospital stay, age, and number of live-born children.
A total of 303 mothers were included in the analysis. Almost two thirds (67.2%) were interviewed at the hospital appointment while 32.8% were interviewed at home. Among these women, 208 (68.6%) had early postpartum discharge and 62 (15.6%) had their discharge delayed for administrative reasons; of these, 14 (6.7%) were in the late discharge group. There were no differences found in age or length of hospital stay between women with delayed discharge due to administrative reasons and those without this delay.
Table I shows sociodemographic and delivery characteristics of participating women, by time of discharge. The mean hospital stay (HS) was 21.5 ± 8.5 hours.
Table II presents symptoms reported up to the seventh day postpartum, analyzed from the time of discharge. From the 303 women included in the analysis, 65% (215) reported at least one symptom after hospital discharge. The most frequent symptom was genital discomfort, which was reported by 26.9% of the women, followed by symptoms suggesting urinary tract infection (23.1%) and symptoms suggesting endometritis (3.4%). We found no significant differences in the occurrence of each of the signs and symptoms by time of discharge (Table II). One woman was hospitalized on the fifth day after delivery with fever.
We fit a multivariate logistic regression model with the dependent variable being the presence of symptoms and the covariates being age of the mother in years, number of live-born children, admission in the second stage of labor, satisfactory prenatal care, application of enema prior to the delivery and having received instructions to have a medical check-up seven days after delivery. The analysis also included the time between maternal discharge from the hospital and the day on which the interview was conducted. The results are shown in Table III.
The raw analysis did not find any variable associated with the presence of early puerperium complications. The analysis adjusted for confounding variables found that women who received no instructions to have a medical check-up seven days after delivery had a higher odds ratio of reporting symptoms of complications than women who received follow-up instructions (OR 1.73; 95% CI 1.01. 2.97). Early discharge was not associated with the presence of symptoms during early puerperium. Nevertheless, in the adjusted models with interaction terms between early discharge and satisfactory prenatal care, we found that women with early discharge and satisfactory prenatal care had a 63% lower odds of presenting symptoms compared with women with late discharge and whose prenatal control was unsatisfactory.
This study found no significant association between early discharge and maternal symptoms of complications during the early puerperium. This result is consistent with other investigators such as Brown et al.12 We found that women who had satisfactory prenatal care and early discharge had lower odds ratio of presenting symptoms of complications than mothers who had early discharge but whose prenatal care was not satisfactory. This study also found that women who received no instructions to get a medical check-up after discharge had a higher risk of presenting complications, what may be an indicator of a positive effect of counseling during the hospital stay.
The association between prenatal care and reported symptoms can be interpreted in the following manner: satisfactory prenatal care serves as an important venue for educating women about postnatal care and their own health;24-26 prenatal visits can help resolve mothers' doubts about events during puerperium as well as provide instructions for when to seek medical care;27,28 early postpartum discharge (EPD) can reduce the window of opportunity for detecting potential complications and for counseling the mother on puerperal care, especially if she did not receive satisfactory prenatal care. It is important to note that even with early discharge, women with satisfactory prenatal care had lower odds of complications than women without early discharge and unsatisfactory prenatal care. This finding indicates that prenatal care plays an important role not only in the prevention of prenatal and delivery complications, but also serves to educate the mother as to early postpartum care.
There are a number of limitations of this study that deserve mentioning. The data on the presence and severity of symptoms of complications during early puerperium were obtained by interviewing the mothers and not by clinical exam or evaluation, which may lead to errors in classification. Also, we did not determine the reasons why mothers did not seek medical attention despite complaining of certain symptoms. However, we presume that the problems in detection and identification of these symptoms, as well as the reasons for not seeking medical care despite the presence of symptoms, are not related to the time of postpartum discharge. This would likely result in a non-differential error in measurement and would only attenuate the associations found between time of postpartum discharge and the occurrence of complications.
Recall bias may be another limitation of this study, as mothers simply tried to remember symptoms when the survey was administered. However, the authors contend that if such bias were present it also would be non-differential, given that mothers were all asked in the same manner (regardless of early or late discharge). Another limitation of the study is the short follow up period of only seven days. Although the majority of the symptoms identified here occur preferentially during the first week postpartum, it is possible that other symptoms or complications appeared after those seven days and were not identified or analyzed, but this issue is part of a future complementary analysis.
In the present study, early discharge was decided by the responsible physician. Women in the early discharge group were all clinically stable. However, for administrative reasons, some women who would have been candidates for early discharge were kept in the hospital. Although it is impossible to control for this effect, we included this "late discharge for administrative reasons" variable in the model and found no significant attributable effect.
The ideal design for evaluating the association between the time of discharge postpartum and the presence of complications would be a randomized clinical trial instead of the cohort observational design used in our study. Such a design would randomize women to early or late discharge groups, and clinical assessment and follow-up would yield more accurate data on puerperal complications. Due to ethical and logistic reasons, it was not possible to use such design in this report.
There may be a residual effect of complications in the immediate postpartum period, which may increase the time to discharge as well as increase the probability of complications appearing during early puerperium. Thus, the results of this study may overestimate the association between early postpartum discharge and the presence and severity of puerperal complications. To decrease the possible bias, we opted for the restriction of the study sample to only those women with no complications at the time of discharge and those with puerperal complications after delivery and before hospital discharge, which could considerably affect the time of postpartum discharge and the probability of later complications.
The results of this study show that although there was no association between early discharge and the severity of complications during early puerperium for all mothers, the presence of symptoms decreased among women who received indications to have a medical check up one week later, and among women with early discharge and satisfactory prenatal care, compared with those with early discharge and unsatisfactory prenatal care, suggesting a positive effect of satisfactory prenatal care even with an early discharge. This correlation deserves further study in order to better understand its importance.
No author has any financial or other relationships that might lead to a conflict of interest. All authors have contributed substantially to the paper. All co-authors contributed to revising and editing drafts of the paper.
We appreciate the support from authorities of the General Hospital of Mexico in conducting this study. We acknowledge Irma Aldana, IIeana Salvador, Araceli Hernandez, Virginia Torres and Emelina Cardoso for their support in conducting the interviews, Noemi Figueroa, Esperanza Piña and Marlene Hernandez for their support in the integration of the database, and Jose Alonso Restrepo for his support in the revision of the manuscript.
1. Olaiz-Fernández G, Rivera-Dommarco J, Shamah-Levy T, Rojas R, Villalpando-Hernández S, Hernández-Avila M et al. Encuesta Nacional de Salud y Nutrición 2006. Cuernavaca, Mexico: Instituto Nacional de Salud Pública, 2006.
2. Grullon KE, Grimes DA. The safety of early postpartum discharge: a review and critique. Obstet Gynecol 1997;90:860-865.
3. Madden JM, Soumerai SB, Lieu TA, Mandl KD, Zhang F, Ross-Degnan D. Health maintenance organization. Effects of a law against early postpartum discharge on newborn follow-up, adverse events, and HMO expenditures. N Engl J Med 2002;347:2031-2038.
4. Braveman P, Egerter S, Pearl M, Marchi K, Miller C. Problems associated with early discharge of newborn infants. Early discharge of newborn and mothers: a critical review of the literature. Pediatrics 1995;96:716-726.
5. Briton JR, Briton HL, Beebe SA. Early discharge of the term newborn: a continued dilemma. Pediatrics 1994;94:291-295.
6. Yaffe MJ, Russillo B, Hyland C. Better care and better teaching. New model of postpartum care for early. Can Fam Physician 2001;47:2027-2033.
7. Lieu TA, Braveman PA, Escobar GJ, Fischer AF, Jensvold NG, Capra AM. A randomized comparison of home and clinic follow-up visits after early postpartum hospital discharge. Pediatrics 2000;105:1058-1065.
8. Norr KF, Nacion K. Outcomes of postpartum early discharge, 1960-1986. A comparative review. Birth 1987;14:135-141.
9. Beck C. Early postpartum discharge programs in the United States: a literature review and critique. Women Health 1991;17:125-138.
10. Helleman LM, Kohl SG, Palmer J. Early hospital discharge in obstetrics. Lancet 1962;1:227-232.
11. Anonymous. Hospital stays continue 10 year decline. Am J Public Health 1992;82:54.
12. Brown SS, Faber RB, Krastev AD. Early postnatal discharge from hospital for healthy mothers and term infants. The Cochrane Database Systematic Reviews 2002, Issue 3, Art. No: CD002958. DOI: 10.1002/14651858.CD002958.
13. Brumfield CG. Early postpartum discharge. Clin Obstet Gynecol 1998;41:611-625.
14. Watt S, Sword W, Krueger P. Longer postpartum hospitalization options who stays, who leaves, what changes? BMC Pregnancy Childbirth 2005;5:13.
15. Madlon-Kay DJ, DeFor TA. Maternal postpartum health care utilization and the effect of Minnesota early discharge legislation. J Am Board Fam Pract 2005;18:307-311.
16. Subdirección General Médica. Jefatura de Servicios de Hospitales. Programa de atención de puerperio de bajo riesgo. Mexico City: Instituto Mexicano del Seguro Social, 1985.
17. de Mucha-Macías RA, Ray LJ, Sánchez NM. Resultados preliminares del puerperio de bajo riesgo en el Instituto Mexicano del Seguro Social. Rev Med IMSS (Mex) 1988;26:333-345.
18. American Academy of Pediatric, American College of Obstetrics and Gynecology. Guidelines for Perinatal Care. 5th ed. Elk Grove Village (IL): American Academy of Pediatrics, 2002.
19. American Academy of Pediatrics. Committee on the fetus and newborn. Hospital stay for healthy term newborns. Pediatrics 1995;96:788-790.
20. Yanover MJ, Jones D, Miller MD. Perinatal care of low-risk mothers and infants. Early discharge with home care. N Engl J Med 1976;294:702-705.
21. Scupholme A. Postpartum early discharge: An inner city experience. J Nurse Midwifery 1981;26:19-22.
22. Avery MD, Fournier LC, Jones PL, Sipovic CP. An early postpartum hospital discharge program. JOGN Nurs 1982;11:233-235.
23. Fajardo-Gutierrez A, Hernandez-Perez A, Huerta-Diaz E, Danglot-Banck C, Gomez-Gomez M, Valle-Gay A. Puerperio de corta estancia hospitalaria. Ginecol Obstet Mex 1996;64:105-109.
24. Norma Oficial Mexicana "NOM-007-SSA2-1993. Atención de la Mujer durante el Embarazo, Parto y Puerperio y del Recién Nacido. Mexico City: Diario oficial de la Federación, Jan 6, 1995.
25. Hosmer DW, Lemeshow S. Applied Logistic Regression. 2nd ed. New York (NY): John Wiley & Sons, 2000.
26. Waldenström U, Rudman A, Hildingsson I. Intrapartum and postpartum care in Sweden: women's opinions and risk factors for not being satisfied. Acta Obstet Gynecol Scand 2006;85:551-560.
27. Gennaro S, Bloch JR. Postpartum health in mothers of term and preterm infants. Women Health 2005;41:99-112.
28. Heaman MI, Gupton AL, Moffatt ME. Prevalence and predictors of inadequate prenatal care: a comparison of aboriginal and non-aboriginal women in Manitoba. J Obstet Gynaecol Can 2005 Mar;27:237-246.
Received on: January 28, 2008
Accepted on: January 26, 2009
This study has been funded by Consejo Nacional de Ciencia y Tecnología, (National Council of Science and Technology). México.
Solicitud de sobretiros: Dolores Ramírez Villalobos. Instituto Nacional de Salud Pública-México (INSP). Av. Universidad 655, 3er piso, Col Sta. Ma. Ahuacatitlán. 62508 Cuernavaca, Morelos, México. E-mail: firstname.lastname@example.org; email@example.com