As if expecting mothers didn’t have enough to worry about, a study has now found that the quantity of delivery complications in hospitals are substantially higher during nights, weekends and holidays, and in teaching hospitals.
Each year, nearly 4m women give birth in US hospitals, making childbirth the most common cause of hospitalisation in this country. Serious but preventable complications occur at the point of delivery, with approximately 700 women dying every year in the U.S. from preventable causes related to pregnancy and childbirth.
In the study, Sammy Zahran, Colorado State University, David Mushinski, Colorado State University, Hsueh-Hsiang Li, Colorado State University, Ian Breunig, Abt Associates Inc, and Sophie McKee, US Department of Agriculture, analysed more than 2m cases from 2005 to 2010, using detailed data obtained from the Texas Department of State Health Services.
The study looked at labour or delivery complications including third- or fourth-degree perineal laceration, ruptured uterus, unplanned hysterectomy, admission to intensive care unit and unplanned operating room procedure following delivery. The research team focused only on women with a single birth (not twins) who had gestation of more than 20 weeks, a delivery attended by a physician, and a normal labour onset.
The study evaluated whether delivery complications (1) vary by work shift (day vs night), (2) increase as the hours pass within work shifts, and (3) increase on weekends (Friday evening to Monday morning) and holidays (Christmas, New Year’s Eve and the Fourth of July).
The results suggest that: the odds of a mother experiencing a delivery complication are 21.3 percent higher during the night shift, and that the odds of a delivery complication increase 1.8% with every hour worked within a shift; a mother delivering an infant on a weekend is 8.6% more likely to encounter a complication than a mother delivering on a weekday; and births occurring on holidays are particularly susceptible to labour or delivery complications, with holiday births being 29.0% more likely to have a complication.
The study also explored whether delivery complication rates are higher in teaching hospitals, and whether they increase when a new cohort of residents enter teaching hospitals in July, causing abrupt declines in physician experience and coordination between members of the healthcare team.
Mothers delivering their infants in teaching hospitals are 2.2 times more likely to experience a delivery complication than mothers birthing at non-teaching hospitals. The risk also increases by a multiplicative factor of 1.3 at teaching hospitals in July, when new residents join the staff rotation. By June, after a full year of training and integration, the risk of a delivery complication at these same hospitals is statistically indistinguishable from chance.
“Across an ensemble of hospital situations where clinical quality is known to vary independently of patient characteristics and volume, we see corresponding variation in the risk of preventable harm to expectant mothers,” states Zahran, an associate professor of demography at Colorado State University.
Obstetric care in hospital settings is a team effort and ineffective teamwork has been implicated in an estimated 75% of preventable medical errors. The researchers hypothesise that hospitals could decrease the risk of harm to mothers by putting more emphasis on scheduling inexperienced physicians with more senior health professionals, among other things.
The establishment of interventions to maximize maternal health requires the identification of modifiable risk factors. Toward the identification of modifiable hospital‐based factors, we analyze over 2 million births from 2005 to 2010 in Texas, employing a series of quasi‐experimental tests involving hourly, daily, and monthly circumstances where medical service quality (or clinical capital) is known to vary exogenously. Motivated by a clinician’s choice model, we investigate whether maternal delivery complications (1) vary by work shift, (2) increase by the hours worked within shifts, (3) increase on weekends and holidays when hospitals are typically understaffed, and (4) are higher in July when a new cohort of residents enter teaching hospitals. We find consistent evidence of a sizable statistical relationship between deliveries during nonstandard schedules and negative patient outcomes. Delivery complications are higher during night shifts (OR = 1.21, 95% CI: 1.18–1.25), and on weekends (OR = 1.09, 95% CI: 1.04–1.14) and holidays (OR = 1.29, 95% CI: 1.04–1.60), when hospitals are understaffed and less experienced doctors are more likely to work. Within shifts, we show deterioration of occupational performance per additional hour worked (OR = 1.02, 95% CI: 1.01–1.02). We observe substantial additional risk at teaching hospitals in July (OR = 1.28, 95% CI: 1.14–1.43), reflecting a cohort‐turnover effect. All results are robust to the exclusion of noninduced births and intuitively falsified with analyses of chromosomal disorders. Results from our multiple‐test strategy indicate that hospitals can meaningfully attenuate harm to maternal health through strategic scheduling of staff.
Sammy Zahran, David Mushinski, Hsueh-Hsiang Li, Ian Breunig, Sophie Mckee