The Texas Health Care Information Collection (THCIC), created by the Texas Legislature in 1995 and part of the Department of State Health Services since 2004, gathers information from hospitals and health maintenance organizations (HMOs) and publishes reports to help people compare and choose their hospitals and health plans. These reports show how Texas hospitals did in 2014, based on indicators of the quality of care for children developed and used by the federal government. This is an annual report on indicators on the quality of care of children in hospitals in Texas.
While most hospitals take care of children, Texas has hospitals in the larger cities that treat only children. These children’s hospitals care for children with complicated conditions or severe injuries, and they have equipment and facilities designed specifically for kids. Staff members in children’s hospital work only with children, and their doctors treat only children. Some of the larger adult hospitals also take care of children. They may have a floor just for children in their hospital or they may send the sicker children to a children’s hospital. However, all licensed general hospitals must be prepared to treat children with emergency medical conditions.
For this report, THCIC says that quality care is: "The degree to which health services for individuals and populations increases the likelihood of desired health outcomes and are consistent with current professional knowledge." In other words, quality care meets the needs of the patient and is based on what works best.
In looking at these pediatric quality indicators, it’s important to remember some key things about these data:
- Children usually are hospitalized less often than adults.
- The pediatric quality measures focus on children from birth to age 18. A procedure done on an infant compared to an older child or teen may be different.
- The reported data come from hospital billing records that include limited information for tracking children’s health.
- Many children need chronic care for conditions like asthma and diabetes, and that care often happens at home or school, or in non-hospital settings like doctors’ offices and urgent care centers. These episodes of care are not included in this report.
- Some of the pediatric quality measures are rarely seen in children, like separation of a surgical wound. The data can make it look like a hospital that had one case of this rare event had poor outcomes; this is a limitations of the measure and not necessarily an indicator of poor quality care.
- Many communities across the state do not have children’s hospitals, and their general hospitals may have pediatric units that treat the more common children's diseases and conditions.
- Some hospitals submit less information to THCIC than others do; complications can only be reported if they were contained in the information sent to THCIC.
Many factors can help you chose a hospital, and these pediatric quality indicators are only one source of information to consider. Other factors that may affect your choice include the specific needs of your child (does he/she need a children's heart doctor or a cancer doctor who only works at some hospitals), the willingness of your insurance to pay different hospitals and doctors, and recommendations from your pediatrician, family and friends. You should use this information to talk with your child’s doctor and hospital, and take a more active role in making health care decisions for your child.
What is the best way to use these reports?
- Familiarize yourself with the six indicators provided in this report.
- Decide which quality indicator is most important for your child, and review that chart. Look to see if the hospital you want to go to is listed. If not, it means that the hospital did not have any cases for that indicator. You may wish to look at another quality indicator for that hospital's performance.
- Rates are calculated for children's hospitals, general hospitals and for the state as a whole. Because children's hospitals may treat children with more complicated conditions or severe injuries, often transferred from other hospitals, it may not be appropriate to compare children's hospitals with general hospitals. However, general hospitals provide good quality care for children and many general hospitals provide specialized care for children.
- Hospitals are arranged alphabetically by metropolitan area. You may wish to compare the performance of hospitals in your area or those that are covered by your health insurance plan. Each chart provides the results for specific hospitals as well as a statewide average and averages by hospital type. Note that guidance is provided for each measure on how to interpret the findings. You may want to look for children's hospitals specifically.
- View the hospital's comments. The hospital name will be followed by (C) in the table if the facility provided comments. It is especially important to view the hospital's comments if performance is lower than expected. Go to the list of comments 2012, 2011, 2010, 2009 or find the hospital in the table of hospital characteristics.
- A table of hospital characteristics (2014) provides the number of licensed beds for each hospital. If the hospital is a teaching facility (one that teaches medical students), it is noted. It is more appropriate to compare the performance of similar sizes and types of hospitals.
- Do not try to compare the results of these reports with other reports you may find on the internet or elsewhere. The results are sure to be different, unless the exact same methodology and the exact same database are used. It is also better to track hospital performance over several years using the same data source.
- Talk to your health care provider.
What do the hospital-specific comparative reports mean?
The Pediatric Quality Indicators, similar to the other quality indicators already available on the THCIC Web site, are measures of health care quality and they are specific to children.
The reports are based on billing data so the results have limitations. Recording billing data–often referred to as “coding”–varies among hospitals. Codes do not tell the whole story of what happened to a patient in the hospital. They are intended for billing not to monitor quality of care, but they are the only data available for reporting on hospitals. The indicators developed by the federal government and used to produce this report adjust for the limitations of the data. When reviewing complication and mortality rates, remember that medicine is not an exact science and complications or death may occur even when all standards of care are followed.
The Texas Legislature directed THCIC to report on how well Texas hospitals do on quality measures. THCIC and its medical and hospital professional advisors agreed that because of the limited options available, the federal Agency for Healthcare Research and Quality (AHRQ) quality indicators are the best ones now available for reports on quality. The data represent the results of hospitals’ treatment of children regarding only these 6 specific procedures or conditions. Since the 2008 data elements were reported, changes will have occurred in the hospital that may impact the quality of care provided. While the data provide a snapshot of a hospital’s performance in treating children, this information should be viewed as one of several resources to help select a hospital provider. These reports only provide some information about hospital performance and are not the only thing you should consider when making a choice.
For many years, the federal government has supported research into what factors affect quality of health care services, including care delivered in hospitals. The federal Agency for Healthcare Research and Quality (AHRQ) has conducted a lot of research. In December 2009, AHRQ released updated software, used to prepare this report, that analyzes billing data and measures certain indicators that relate to outcomes of care provided within hospitals: volume and mortality for a specific procedures, and rates for specific complications that occurred during stays in the hospital.
Are the comparisons between hospitals appropriate?
Because of their expertise, some hospitals treat more high-risk patients. And, some patients arrive at hospitals sicker than others. That makes comparing the quality of care at different hospitals hard. This is also complicated by comparing general hospitals that treat children and adults with children’s hospitals that treat only kids. The information in this report tries to take into account differences among hospitals, in the age of their patients and how sick they are. Detailed information about the process used to organize and adjust the data for study purposes can be found on the AHRQ web site.
The performance of the hospitals in this report may reflect if the care ordered by a doctor and delivered at a given hospital was consistent with standards of care. However, it also may reflect factors that do not relate to hospital performance, such as patient or physician choice, how sick the child was, age, other illnesses at the time, or the availability of specialized equipment or doctors. The data analysis method attempts to adjust for some of these factors. Results published here should be considered along with other factors when you are trying to make choices.
You should remember that doctors direct and oversee the medical care that is delivered at hospitals, and order tests, medicines and treatments. This report does not separate the effect of the doctor from the effect of the hospital. The quality of care provided in a hospital comes from how well its doctors, nurses, support staff and management work together as well as the technology and other resources available in the hospital. If a major change occurs that affects any of these–such as the departure of a key surgeon or the addition of new technology–the indicators may change quickly. As you talk with your doctor and hospital about your child’s care, ask questions about what changes, if any, have occurred that could affect the care your child receives. Medical practice and standards of care change over time as new technology and medicines become available, and as research studies show what might work the best.
Where do the THCIC data come from?
The data collected by THCIC come from information that hospitals record primarily for billing purposes. This type of record, referred to as "administrative data," consists of codes for diagnoses and procedures along with information about the patient's age, gender, accompanying medical conditions and discharge status. While administrative data should not be used as the only source of information on health care quality, it can provide an idea of the medical care being delivered by hospitals. The THCIC data reflect only the care provided to patients who were discharged from the hospital in calendar year 2008. No patient and physician identifying information is available to preserve confidentiality. Also, to protect patients and physicians, when fewer than 5 patients in a facility had a specific procedure, the number of procedures is not included in the report. If a hospital had fewer than 30 patients with a specific procedure, no comparative rates were calculated because the results would not be reliable.
After THCIC checks the data for errors, hospitals are given a short time to review and fix mistakes in the data they supplied. Hospitals are also able to review this report before its release. Hospital comments regarding their performance on these indicators are included in the report and are available on THCIC's Web site.
Most rural hospitals are not included in this report because state law does not require them to report data to the THCIC.
Definitions of terms used in the charts and tables of data
- Confidence Interval - The confidence interval is like a margin of error and takes into account the possibility that a hospital's performance could be affected by random chance. Another way of explaining the confidence interval is that there is a 95% chance that the actual risk-adjusted mortality or complication rate is within the confidence interval. The range will vary for each hospital depending on the number of cases for that condition or for that procedure, and the standard error rate for the year.
If the confidence interval range is below the state average, the hospital's rate is significantly lower than the state average. If the confidence interval range is above the state average, the hospital's rate is significantly higher than the state average. If the state rate is within the hospital’s confidence interval, the hospital's rate is not significantly different from the state average.
- Expected Rate - The hospital's expected rate is calculated by applying a national average hospital rate to the hospital's case mix. This rate reflects an expectation of the hospital's mortality or utilization rates if it had performed at the level of the national average.
- Metropolitan Statistical Area (MSA) - Geographic area consisting of a large population city together with nearby communities that are linked with the city.
- Observed Rate - The number of reported events, such as the number of complications or deaths, for the heart surgery indicator, divided by the total number of patients who could have had the event. Observed rates are simple measures of performance. When compared to the risk-adjusted rates, the effect of patient case mix on that hospital's performance can be seen.
- Risk-adjusted rate - Adjustments made to the THCIC data based on patient characteristics such as age, gender and medical codes (diagnostic groups) for a specific condition or procedure. The risk-adjusted rate is the best guess of what the hospital's rates would have been if the hospital had a mix of patients identical to a national-average patient mix for the year.
Hospitals with significantly lower risk-adjusted mortality or complication rates than observed rates may have a more complex patient case mix. Hospitals with significantly higher risk-adjusted rates than observed rates may have a less complex patient case mix.
Technical Specifications for the Indicators
Inclusion and exclusion criteria for each indicator: