Racial and ethnic disparities in patient safety events for postoperative complications

Figure 1 shows the relative rates for minority groups compared to whites of complications and problems that result from having an operation (i.e., postoperative complications). Except for postoperative wound dehiscence (when the surgical wound opens after an operation on the stomach or pelvis), minority groups generally had higher rates than whites for postoperative complications. In fact, all three minority groups had 21 to 34 percent higher complication rates than whites for postoperative physiologic and metabolic derangement (abnormal changes in body function after an operation) and 14 to 21 percent higher rates for postoperative respiratory failure (when a patient is unable to breath after surgery and temporarily needs a ventilator). Blacks and Asian-Pacific Islanders had 18 to 21 percent higher rates of postoperative hemorrhage or hematoma (when a patient bleeds too much or develops a large blood clot) and 25 to 42 percent higher rates of postoperative sepsis (a serious bloodstream infection) than whites. In contrast, Hispanics had a 13 percent lower rate than whites for postoperative hemorrhage or hematoma.

Figure 1

Rates of patient safety events for postoperative (PO) complications among minorities relative to whites, risk adjusted, 2005. Source: Agency for Healthcare Research and Quality, Center for Delivery, Organization, and Markets, Healthcare Cost and Utilization (more...)

Racial and ethnic disparities in patient safety events for medical and other surgical complications

Figure 2 indicates the relative rates for minority groups compared to whites of medical and other surgical complications and problems that occurred during a hospital stay. Except for death in low-mortality DRGs (medical and surgical conditions that rarely result in death), where all minority groups had better outcomes than whites, the rates for these complications among minority groups were generally comparable to the complication rates among whites. However, there were three indicators in which one minority group had substantially higher rates than whites. The largest disparity occurred for blacks, who had a rate of selected infections due to medical care that was 29 percent higher than the rate among whites. Asian-Pacific Islanders were 16 percent more likely than whites to have a failure to rescue (when a patient dies after developing a complication that should have been identified quickly and treated during hospitalization). Moreover, the rate of accidental puncture or laceration was 12 percent higher for Asian-Pacific Islanders compared to whites.

Figure 2

Rates of patient safety events for medical and other surgical complications among minorities relative to whites, risk adjusted, 2005*. *Rates are observed rates and are not adjusted for age, gender, or diagnostic risk factors. Source: Agency for Healthcare (more...)

It is also noteworthy that Hispanics had a 22 percent lower rate than whites for iatrogenic pneumothorax (when air leaks out of the lungs because it was accidentally punctured).

Racial and ethnic disparities in patient safety events for obstetrical complications

Figure 3 denotes the relative rates for minority groups compared to whites of complications that arise during childbirth. In general, both blacks and Hispanics had better, i.e., lower, obstetrical complication rates when compared to whites, while Asian-Pacific Islanders had worse outcomes for most obstetrical complications.

Figure 3

Rates of patient safety events for obstetrical complications among minorities relative to whites, risk adjusted, 2005*. *Rates are observed rates and are not adjusted for age, gender, or diagnostic risk factors. Source: Agency for Healthcare Research (more...)

Compared to whites, blacks and Hispanics had consistently lower rates of injury to the neonate during birth (e.g., broken collarbone, an infection, or a head injury) and obstetrical traumas (i.e. perineum tears) occurring during vaginal delivery with and without instruments (e.g., forceps or other medical instruments). Such tears, which can happen even when medical instruments are not used, are often preventable. Moreover, the rate of obstetrical trauma occurring during Cesarean delivery was also lower in Hispanics than the rate experienced by whites.

Conversely, the rates of obstetrical traumas among Asian-Pacific Islanders were 30 to 40 percent higher than those for whites. However, Asians-Pacific Islanders and whites had similar rates of injury to the neonate.

Patient Safety Indicators

The AHRQ Patient Safety Indicators (PSI) (version 3.1) were used for this analysis to identify the admissions of interest and for risk-adjustment. The PSIs, a component of the AHRQ QIs, are a set of measures that can be used with hospital inpatient discharge and administrative data to provide a perspective on quality. PSIs screen for problems that patients experience as a result of exposure to the healthcare system and that are likely amenable to prevention by changes at the system or provider level. These are referred to as complications or adverse events.

PSIs are defined on two levels: the provider-level and the area-level. Provider-level indicators provide a measure of the potentially preventable complication for patients who received their initial care and the complication of care within the same hospitalization. Provider-level indicators include only those cases where a secondary diagnosis code flags a potentially preventable complication. Individual provider-level PSIs were included in the Statistical Brief if the SID disparities analysis file could provide reliable estimates for all four racial/ethnic groups (i.e. there was a large enough sample size) and had strong evidence that it measures a hospital patient safety event. This Statistical Brief includes provider-level measures for hospital admission rates for the following 14 potentially preventable hospital complications and iatrogenic events:

• PSI 2: Death in low-mortality DRGs – How often patients died in the hospital when they had been admitted for a health problem that rarely results in death. (This is called death in low mortality DRGs, i.e. diagnosis-related groups.)
• PSI 4: Failure to rescue – How often patients died after developing a complication that should have been identified quickly and treated (called failure to rescue).
• PSI 6: Iatrogenic pneumothorax – How often air leaks out of the patient’s lung because someone accidentally punctured during a medical procedure or operation (a complication called iatrogenic pneumothorax).
• PSI 7: Selected infections due to medical care – How often patients got certain types of infections as a result of the care they received in the hospital.
• PSI 9: Postoperative hemorrhage or hematoma – How often patients bled too much (called hemorrhaging) or developed a large blood clot after an operation (which is called a hematoma).
• PSI 10: Postoperative physiologic and metabolic derangement – How often hospital patients experienced problems with blood sugar control (if they have diabetes) or kidney failure (if they did not have previous kidney trouble) after having an operation (these problems are called postoperative physiologic and metabolic derangements).
• PSI 11: Postoperative respiratory failure – How often patients became unable to breathe on their own following an operation, and needed a ventilator (a machine that helps someone breathe), at least temporarily. (This is called postoperative respiratory failure.)
• PSI 13: Postoperative sepsis – How often hospital patients got a serious bloodstream infection following an operation (which is called postoperative sepsis).
• PSI 14: Postoperative wound dehiscence – How often a surgical wound in the stomach or pelvic area split open after an operation. (This is called postoperative wound dehiscence in abdominopelvic surgical patients.)
• PSI 15: Accidental puncture or laceration – How often patients were accidentally cut, making an unnecessary or dangerous hole or tear in an organ of the body (called an accidental puncture and laceration), while receiving medical care in the hospital.
• PSI 17: Birth trauma-injury to neonate – How often a newborn infant experiences a problem during the birth process (labor or delivery) such as a broken collarbone, an infection, or a head injury.
• PSI 18: Obstetric trauma-vaginal delivery with instrument – How often a woman experiences a tear (trauma) to her perineum—the area between her vagina and rectum—while giving birth, when a health care provider is helping to deliver her baby using a forceps or other medical instrument.
• PSI 19: Obstetric trauma-vaginal delivery without instrument – How often a woman experiences a tear (trauma) to her perineum—the area between her vagina and rectum—while giving birth.
• PSI 20: Obstetric trauma-Cesarean delivery – How often a woman experiences a tear (trauma) in her perineum—the area between her vagina and rectum—or to any of the birth-related organs inside her body, during a Cesarean (surgical) delivery of a baby.

Further information on the AHRQ QIs, including documentation and free software downloads, is available at http://www.qualityindicators.ahrq.gov/index.htm.

Risk adjustment

Except for the death in low mortality DRGs and birth trauma-Cesarean delivery indicators, which describe the observed rate, the PSI rates used in this Statistical Brief were adjusted for select risk factors. The PSI risk adjustment variables were age, gender, age-gender interaction, DRG cluster, and comorbidity. Regression-based standardization (designed by the developers of the PSI software) was used for risk adjustment.

The rate calculations also take into account the clustering of patients within hospitals and the hospital stratification aspects of the sampling design for the SID disparities analysis file. The patient safety event rates were weighted for national estimates. Pairwise t-tests were used to test for statistical significance of differences in patient safety (additional detail on the file development, risk adjustment methodology, and statistical approach is available in Coffey et al., 2007).

Reporting of race and ethnicity

Race and ethnicity measures can be problematic in hospital discharge databases. Some states do not collect information on race/ethnicity from hospitals, and within states that collect the information, some hospitals do not code race and ethnicity reliably. The 2005 SID disparities analysis file was limited to 23 states, and to hospitals within those states with good reporting of race and ethnicity.

Data on Hispanics are collected differently among the states and also can differ from the Census methodology of collecting information separately on race (white, African American, Asian, American Native) and ethnicity (Hispanic, non-Hispanic). State data organizations often collect Hispanic ethnicity as one of several categories that include race. Therefore, for multistate analyses, HCUP creates the combined categorization of race and ethnicity for data from states that report ethnicity separately. When a state data organization collects Hispanic ethnicity separately from race, HCUP uses Hispanic ethnicity to override any other race category to create an Hispanic category for the uniformly coded race/ethnicity data element, while also retaining the original race and ethnicity data. This Statistical Brief reports the HCUP uniform coding of race/ethnicity for the following categories: white, non-Hispanic; black, non-Hispanic; Asian-Pacific Islanders, non-Hispanic, and Hispanic.

Relative rates

Adjusted relative rates are calculated by dividing the minority group adjusted patient safety event rate by the non-Hispanic white adjusted rate. Minority groups that have higher rates than non-Hispanic whites have a relative rate greater than 1.0, while those with patient safety event rates lower than non-Hispanic whites have a relative rate less than 1.0. A relative rate equal to 1.0 describes rates among minority groups and non-Hispanic whites that are similar. A relative rate significantly less than 1.0 or greater than 1.0 may signal disparities in the quality and safety of hospital care.

Types of hospitals included in HCUP

HCUP is based on data from community hospitals, defined as short-term, non-Federal, general and other hospitals, excluding hospital units of other institutions (e.g., prisons). HCUP data include OB-GYN, ENT, orthopedic, cancer, pediatric, public, and academic medical hospitals. They exclude long-term care, rehabilitation, psychiatric, and alcoholism and chemical dependency hospitals, but these types of discharges are included if they are from community hospitals.

Unit of analysis

The unit of analysis is the hospital discharge (i.e., the hospital stay), not a person or patient. This means that a person who is admitted to the hospital multiple times in one year will be counted each time as a separate “discharge” from the hospital.