Ms. Hiebert-White is executive editor, Health Affairs.
Although Congress and other public policymakers have debated endlessly on the financing and restructuring of America's healthcare system, most are loath to micromanage medical practice decisions — especially in this current era of Republican leadership. Yet this summer a number of state and congressional leaders have begun to propose and pass legislation that would dictate the length of a hospital stay for childbirth. This is in response to a perception that insurers are mandating very early hospital discharge after childbirth for economic reasons at the risk of potential adverse medical outcomes to infants and mothers.
This debate of economics versus medical quality raises the question of who is driving medical decision making — healthcare professionals, public policymakers, or payers. How much power should insurers have in setting medical care standards? Who is responsible for proving medical efficacy of a given course of treatment? This column examines these questions, as well as the current activities of state and federal policymakers and of provider groups in relation to hospital length of stay for childbirth.
Shorter Stays, Growing Concern
In the United States, hospital stays after childbirth have decreased dramatically through the years. A recent study by the Centers for Disease Control and Prevention reported that the median length of stay after vaginal birth decreased by 46 percent from 1970 to 1992 (3.9 days to 2.1 days). For cesarean delivery, the median length of stay during the same period fell 49 percent (from 7.8 days to 4 days).1 These statistics include labor times and complicated deliveries, so the postpartum lengths of stay for uncomplicated births are likely to be shorter.
In recent years these lengths of stay have fallen even more dramatically. The American Medical Association (AMA) Council on Scientific Affairs reported to the AMA delegation in June 1995 that "in the last three or four years typical hospital stays have dwindled to 24 hours or less for uncomplicated vaginal deliveries and two to three days for Cesarean deliveries."2
In 1992 the American Academy of Pediatrics (AAP) and American College of Obstetricians and Gynecologists (ACOG) issued a joint set of guidelines for perinatal care. These provider groups called for lengths of stay of 48 hours for vaginal delivery and 96 hours for cesarean birth, when no complications are present and excluding the day of delivery. The AAP/ACOG guidelines warn:
Because many neonatal problems do not become apparent until several days after birth, there is an element of medical risk in early neonatal discharge. Although most problems are manifest during the first 6 hours, data suggest that readmissions may be more common when early (by 48 hours) or very early (by 24 hours) discharge programs are instituted.3
In May 1995 ACOG issued a new statement outlining its concern about lengths of stay that routinely fall below the 1992 guidelines:
The recent trend to even shorter length of stay following delivery appears to be driven primarily by financial motivations. . . . At a time when obstetrical delivery is the most frequent cause of hospitalization in the United States, the shortening of a woman's hospital stay holds obvious appeal to insurers.4
An ACOG representative said the physicians' group was troubled by insurers shortening covered stays without empirical evidence to support this action. At press time, a representative of the AAP said the group was also preparing a new policy statement to be released "imminently."
The AMA placed its concern regarding shorter hospitalizations on the record at its June 1995 annual delegates meeting. The association adopted a statement regarding:
- Its concern about shorter stays "in the absence of data to demonstrate safety"
- Its concern that "discharge should be based on the attending physician's clinical judgement and not economic considerations"
- Follow-up screening procedures for infants
- Encouragement of "well-designed studies to identify safe postpartum stay hospital discharge practices
- Support of efforts by the Maternal and Child Health Bureau (MCHB) of the Health Resources and Services Administration and other groups to examine the issue of appropriate medical care
The AMA report recognizes that early discharge can reduce the mother's and baby's risk of infection from being in the hospital. However, it cautions that the current lengths of stay "may not provide adequate time for routine medical and social assessment" of mother and child.
Managed care plans counter that it is important to look at the continuity of care and what is best for each patient, not one-size-fits all lengths of stay. "For a patient who needs to be hospitalized [beyond a minimum stay], they are covered. There's a difference between guidelines, targets, and what's covered. That's getting lost in debate," said Susan Pisano, spokesperson for the Group Health Association of America (GHAA), in an interview. GHAA is the nation's largest and oldest trade group for health maintenance organizations (HMOs).
"We don't think it's a good idea to legislate clinical practice; the decision should be made by the physician in the best interest of each patient," Pisano continued. "We think most physicians would agree that once it is safe to leave the hospital, the home becomes a superior environment in which to be a family. When it's safe to be home, you don't want to be in the hospital."
Quality of Care: Whose Burden of Proof?
The consumer movement of the 1970s and 1980s initially drove down lengths of stay as mothers pushed for less "medicalized" and more family-centered approaches to childbirth. The most recent trends, however, appear to be driven by economics. With 3,870,000 births in 1993, up 10.9 percent from 1983 (according to recent American Hospital Association data), efforts to contain costs can provide substantial savings for insurers.5
Childbirth represents a major part of the care provided at hospitals; in 1992 "females with deliveries" made up 12.6 percent of all short-stay hospital discharges, according to data from the National Center for Health Statistics.6 Although heart disease, when grouped to include numerous diagnoses, made up a slightly larger percentage of hospital discharge (12.7 percent), deliveries represented the single largest diagnostic category for hospitalization.
Both fee-for-service and managed care plans are pushing the envelope on lengths of hospitalization for childbirth, with 24 hours routine for vaginal deliveries and 12- and 6-hour stays being considered in some parts of the country. Provider groups are questioning whether insurers who are pushing these changes should be responsible for showing their safety and efficacy.
The AMA's June 1995 report asks: "Where should the burden of proof lie? Is it sufficient that a change in practice fail to create significant adverse consequences, or should it be demonstrated to be 'safe'? . . . Evidence that early discharge is not unsafe does not mean it is safe." The statement adopted by the AMA delegation recommends that "in the absence of definitive empirical data, perinatal discharge of mothers and infants should be determined by the clinical judgement of attending physicians and not by economic considerations."
ACOG's May 1995 statement clearly places the burden on insurers:
ACOG believes that changes in practice . . . should be based on sound scientific data that demonstrate good outcomes for mother and infant, as well as being cost effective. As yet, these data do not exist. Until they do, the burden of proof of safety of early discharge rests with those who are driving the change.
GHAA's Pisano responded to the ACOG challenge: "We are [now] writing to ACOG and offering to share data with them. It's unproductive to have this fought out in the media. There's actually a lot of common ground here, and it would be more productive if we did develop a process for sharing information." Pisano said HMOs are particularly well placed to track and respond to data on medical practice, given their organizational structure and incentive to look at the long-term health of their members.
Research Limitations As noted by the AMA and ACOG statements, a key problem is the lack of clear scientific evidence proving the safety or harm of shorter lengths of stay. A recent literature review commissioned by MCHB conducted by University of California, San Francisco (UCSF), researcher Paula Braveman concluded: "Although many studies have examined early discharge, when standard scientific criteria are applied, it becomes clear that the currently available literature provides little scientific knowledge to guide discharge planning for apparently well newborns and their mothers."7
Some of the research limitations include inadequate statistical power, lack of random assignment, careful participant selection, and shifting definitions of "early" discharge and "good" outcome. Most early discharge programs studied to date have been voluntary and have not examined the effect of early discharge on disadvantaged, high-risk populations.
In December 1994 MCHB convened a group of provider organizations, academics, and other experts to examine the issue of increasingly shortened hospitalizations for childbirth. The group agreed that "concentrated work on developing national standards for adequate and sufficient care during the early neonatal period" is needed. In addition, the group decided that, rather than discuss the precise number of hours for length of stay, it would be more productive to "work to identify the essential prerequisites for timely and appropriate perinatal discharge."
One HMO system that has publicly responded to the debate is Kaiser Permanente's Northern California Region. In a June 29 press statement, the regional system announced it was launching a multiyear study of maternal and infant outcomes after discharge from the hospital to be conducted by UCSF's Braveman and Drs. Tracy Lien and Gabriel Escobar of Kaiser Permanente's Division of Research. Kaiser Permanente also reported that a 1994 study of 19,000 births in Northern California found "no difference in the rate of re-hospitalization for babies discharged earlier than 24 hours and later than 24 hours after delivery." The group's 1994 data (some of the most recent publicly available) show that 60 percent of new mothers in Kaiser's Northern California region go home after 24 hours, and 40 percent go home within 24 hours.
With scientific research currently lacking, however, the recent attention on this issue is driven in part by anecdotal reports of infants experiencing severe problems — even death — because of early discharge. Doctors and hospitals are reporting that infants have problems such as untreated jaundice, dehydration because of problems with breastfeeding, and blood poisoning from Streptococcus B infection.
Doctors at Children's Hospital Medical Center in Cincinnati reported a 30 percent increase in infant readmissions for jaundice and three times more readmissions for severe dehydration in babies under four weeks of age between 1992 and 1994.8 During this period, the average length of stay for normal delivery fell from 72 hours to 24 hours in Cincinnati.
I had the opportunity this past winter to compare how the U.S. and French healthcare systems handle childbirth. I gave birth to my first daughter in January after an arduous, 45-hour labor and finally a cesarean section. My fee-for-service health plan allowed three days in the hospital. My final bill, including prenatal care, totaled more than $13,000; out-of-pocket costs were about $3,000.
My sister delivered her second son just 24 hours before I gave birth. Her delivery took place at a hospital in the south of France. French Social Security allowed a six-day stay for her normal, uncomplicated delivery, though a five-day stay is average. The national insurance program covered virtually the entire bill of 9,418 francs, or a little more than $1,700, including prenatal care. My sister's out-of-pocket costs were 88 francs, or about $16 (for telephone calls and for three extra television stations).
We both had private rooms. We had technologically similar care. For instance, my sister had four prenatal ultrasounds (two to three is the average in France, with ultrasounds mandated at five months' and eight months' gestation). I, on the other hand, had two ultrasounds, which were covered by insurance only because I could prove medical necessity. When I expressed concern to the technician regarding whether my insurance carrier would cover the second ultrasound, he changed the charge code to a less expensive designation, in case I was saddled with the bill out of pocket.
In my experience, I observed points where my treatment was driven by insurance coverage and service constraints to contain cost. My three-day stay for cesarean delivery fell short of the AAP/ACOG guidelines because of insurance restrictions. My insurer questioned numerous items on my bills. I saw how codes are changed depending on whether providers think they will be covered by insurance or paid by the patient. The hospital nursing staff complained to me about drastic nursing layoffs related to insurers' pressure to lower lengths of stay, and I experienced long waits for care as a result of understaffing. A promised lactation specialist never showed up during my three-day stay.
My sister experienced none of these restrictions. Instead, she attended child-care classes while staying at the maternité, enjoyed nursing support for breastfeeding, and had her son monitored for healthy neonatal development. The French health system even covered 10 postnatal sessions with a midwife in private practice to monitor my sister's recovery and help her regain control of her perineal and abdominal muscles through physical therapy.
While I acknowledge the anecdotal nature of my experience, it did raise questions regarding what really drives medical practice. In my case, differing insurance coverage standards played a major role. When the price tag for childbirth in the United States is dramatically more expensive than delivery in other countries, one can see why insurers are eager to control treatment costs. It makes me wonder why our costs are so high. Have our cost shifting, perverse financial incentives, and high-tech capitalization priced U.S. childbirth into the stratosphere? Does France underpay its hospitals and physicians for the true cost of care? Where does the reality of best medical practice lie?
Public Policymakers Legislating Medical Practice
Recently, some public policymakers have taken up the debate about short hospital stays for childbirth. Sen. Bill Bradley, D-NJ, introduced legislation June 27, 1995, to allow minimum stays of 48 hours for births and 96 hours for cesarean deliveries, citing concern about "drive-through deliveries." His bill, cosponsored by Sen. Nancy Kassebaum, R-KS, chairperson of the Senate Labor and Human Resources Committee, specifies that shorter stays are permitted (if neither the mother nor attending physician object) and calls for home healthcare to be provided after shorter stays. At press time, hearings on the legislation were anticipated before the August recess.
In the House, Rep. Bernard Sanders, Independent-VT, proposed a nonbinding resolution (H.Con.Res. 79) June 27 to recommend minimum stays for childbirth in accord with AAP/ACOG guidelines. Language identical to the resolution was offered by Rep. John Edward Porter, R-IL, and was accepted unanimously as an amendment to the fiscal year 1996 House Labor, Health and Human Services, and Education appropriations bill.
Sanders, along with six Democratic congressional colleagues, delivered a letter August 2 to House Commerce Committee Chairperson Rep. Thomas Bliley, Jr., R-VA, calling for urgent hearings on the issue. The members write: "None of us believes that Congress should micromanage insurance practices, but neither do we believe that insurance companies should be able to dictate medical decisions. We need hearings to shed some light on this critical health care issue to assess the need for legislative action."
Two states, New Jersey and Maryland, recently passed legislation setting limits on early discharge for childbirth. Maryland's law, HB 888, requires insurers to follow the AAP/ACOG standards, which allow early release if certain criteria are met or the insurer agrees to cover postpartum home healthcare visits. The bill was signed into law May 25 by Democratic Gov. Parris Glendening and goes into effect October 1.
The New Jersey law, AB2224, requires 48-hour stays for normal delivery and 96 hours for cesarean. Shorter stays of 24 hours are allowed if the insurer covers three home healthcare visits. The law passed the New Jersey legislature unanimously and was signed by Republican Gov. Christine Todd Whitman on June 28. That such a law would be signed by a governor who has built her reputation as a cost cutter is noteworthy. Other states that have introduced legislation for childbirth hospital stays include Illinois, Pennsylvania, and New York.
This type of legislation to hold the line against insurers' efforts to cut costs clearly enjoys bipartisan support. Yet it is ironic that the very policymakers who want to cut federal and state healthcare costs are drawing this legislative line in the sand on the medical practice surrounding childbirth. What does it mean to have medical practice guidelines legislated at this level? Is this really the responsibility of government? Or is it a reaction to the perception that the cost competition driving the healthcare sphere today has gone too far? In all this, what role is left for the healthcare providers — hospitals and physicians — in defining what is best medical practice?
- Centers for Disease Control and Prevention, "Trends in Length of Stay for Hospital Deliveries — United States, 1970-1992," Morbidity and Mortality Weekly Report, May 5, 1995, pp. 335-337.
- American Medical Association, "Impact of 24-Hour Postpartum Stay on Infant and Maternal Health" (Resolution 135, A-94), Council on Scientific Affairs Report 5 - A-95, 1995.
- American Academy of Pediatrics and American College of Obstetricians and Gynecologists (ACOG), Guidelines for Perinatal Care, 3d ed., 1992, pp. 105-111.
- ACOG, "Statement on Decreasing Length of Hospital Stay Following Delivery," May 23, 1995.
- American Hospital Association, Hospital Statistics, 1994-95 ed., Chicago, 1994.
- National Center for Health Statistics, National Hospital Discharge Survey: Annual Summary, 1992, Vital and Health Statistics, October 1994.
- Paula Braveman, "Early Discharge of Newborns and Mothers: A Critical Review of the Literature," unpublished report commissioned by Maternal and Child Health Bureau, Department of Health and Human Services, December 1994.
- Sandra G. Boodman, "Discharged Too Soon?" Washington Post, Health Section, June 27, 1995.
Copyright © 1995 by the Catholic Health Association of the United States
For reprint permission, contact Betty Crosby or call (314) 253-3477.