Findings (continued)

Long-Stay and Expensive and Procedures

Which Procedures Are Associated with the Longest Hospital Stays?

• Bone marrow and other organ transplantations (e.g., liver, heart) are associated with some of the longest hospital stays.
• Other procedures associated with long hospital stays are indicative of patients with serious chronic illnesses. Ileostomy, gastrostomy, and enteral/parenteral nutrition are performed for very ill patients requiring extensive care. Tracheostomy is performed for patients who are often dependent on mechanical ventilation or have need for other types of extended care.
• The hospitalizations with the longest lengths of stay are not very common. Collectively, these 10 procedures represent less than 2 percent of all hospital stays.

Select Figure 3 (12 KB), Procedures Associated with Longest Hospital Stays.

Which Procedures Are Associated with the Most Costly Hospital Stays?

• Four of the top 10 most costly hospital stays are related to procedures on the cardiovascular system:
  • Heart transplantation (part of other organ transplantation).
  • Heart valve procedures.
  • Other operating room (OR) heart procedures (e.g., implant of pulsation balloon, pericardiotomy).
  • Other vascular bypass and shunt (e.g., intra-abdominal venous shunt, aorta-renal bypass, aorta-subclavian-carotid bypass).
• Bone marrow and organ transplantations (e.g., heart, liver, kidney) are associated with some of the most expensive hospital stays.
• Some procedures are not in themselves costly but are associated with critical illness and expensive care, for example, a long stay in an intensive care unit. These include:
  • Tracheostomy.
  • Swan-Ganz catheterization.
  • Other OR respiratory procedures (e.g., reopening of recent thoracotomy site, decortication of lung, scarification of pleura).
• Four of the procedures performed during the most expensive hospital stays also are among those with the longest lengths of hospitalization:
  • Other organ transplantation (e.g., heart, liver).
  • Tracheostomy.
  • Bone marrow transplantation.
  • Ileostomy/enterostomy.

Select Figure 4 (15 KB) Procedures Associated with Most Costly Hospital Stays.

Return to Contents

In-Hospital Deaths

Which Procedures Are Performed Most Often during Hospital Stays Ending in Death?

• The most common procedure performed during hospital stays that end in death is conversion of cardiac rhythm, which indicates an unsuccessful attempt at resuscitation.
• The second most common procedure associated with high in-hospital mortality is respiratory intubation and mechanical ventilation. This procedure is done for critical conditions such as respiratory failure, myocardial infarction, stroke, pneumonia, and septicemia.
• Procedures that indicate the presence of organ failure and critical illness include Swan-Ganz catheterization, enteral and parenteral nutrition, hemodialysis, and blood transfusions.
• It is important to note that these top 10 procedures are markers of severe underlying disease and are not the cause of death.

Table 4. Procedures Associated with Highest Inpatient Mortality

All-Listed Procedures*	In-hospital Mortality (in Percent)
1. Conversion of cardiac rhythm	39.2
2. Respiratory intubation and mechanical ventilation	31.1
3. Swan-Ganz catheterization for monitoring heart function	25.2
4. Enteral and parenteral nutrition	16.9
5. Incision of pleura, thoracentesis, chest drainage	12.7
6. Diagnostic bronchoscopy and biopsy of bronchus	12.7
7. Blood transfusion	9.1
8. Hemodialysis	8.7
9. Computerized axial tomography (CT) scan, head	7.0
10. Upper gastrointestinal endoscopy, biopsy	4.2

* The term "All-listed procedures" refers to all procedures performed during a hospital stay; patients often receive more than one procedure.

Return to Contents

Charges

What Are the Differences in How Procedures Are Billed, by Body System?

• Obstetrical procedures comprise 38 percent of all procedures billed to Medicaid, 28 percent of those billed to private insurers, and 22 percent of those performed during uninsured hospital stays.
• Cardiovascular procedures account for more than 2 in 5 of all procedures billed to Medicare, 1 in 7 of those billed to Medicaid, and 1 in 5 of those billed to private insurers or uninsured.
• Procedures of the digestive system are the second most common category of procedures for Medicare patients, performed on more than 1 in 5 patients.

Select Figure 5 (17 KB), How Procedures Are Billed, by Body System.

What Are the Most Common Procedures for Hospital Stays Billed to Medicare?

• Five out of the top 10 procedures billed to Medicare are related to the heart: diagnostic cardiac catheterization, CABG, cardiac pacemaker or cardioverter/defibrillator procedure, diagnostic ultrasound of heart, and PTCA.
• Medicare is billed for about three-fourths of all hospital stays that involve a cardiac pacemaker or defibrillator procedure.
• Medicare is billed for over 65 percent of all hospital stays that involve hemodialysis (treatment for renal failure) because treatment for end-stage renal disease is covered by Medicare, regardless of the patient's age.
• Medicare is billed for roughly half of all hospital stays involving cardiac catheterization, blood transfusion, upper gastrointestinal endoscopy, CABG, diagnostic ultrasound of heart, and PTCA.

Table 5. Most Common Procedures for Hospital Stays Billed to Medicare

All Listed Procedures*	Total number of discharges with this procedure category (in thousands)	Medicare's share of hospital stays with this procedure (in percent)
1. Diagnostic cardiac catheterization, coronary arteriography	1,954	50.5
2. Respiratory intubation and mechanical ventilation	701	44.8
3. Blood transfusion	696	53.7
4. Upper gastrointestinal endoscopy, biopsy	651	55.7
5. Coronary artery bypass graft (CABG)	395	53.4
6. Cardiac pacemaker or cardioverter/defibrillator procedure	349	75.1
7. Colonoscopy and biopsy	343	60.7
8. Diagnostic ultrasound of heart (echocardiogram)	339	51.9
9. Hemodialysis	338	65.5
10. Percutaneous transluminal coronary angioplasty (PTCA)	296	48.7

* The term "All-listed procedures" refers to all procedures performed during a hospital stay; patients often receive more than one procedure.

What Are the Most Common Procedures for Hospital Stays Billed to Medicaid?

• Seven of the top 10 procedures billed to Medicaid are for pregnancy, childbirth, and newborn infant care. Overall, Medicaid is billed for about one-third of all hospital stays involving these childbirth procedures.
• Medicaid is billed for 1 in 5 hospital stays involving respiratory intubation and mechanical ventilation.
• Medicaid's share of hospital stays involving diagnostic cardiac catheterization—7.2 percent—is less than half its share of stays requiring blood transfusions—15.2 percent.

Table 6. Most Common Procedures for Hospital Stays Billed to Medicaid

All Listed Procedures*	Total number of discharges with this procedure category (in thousands)	Medicaid's share of hospital stays with this procedure (in percent)
1. Procedures to assist delivery	752	36.6
2. Fetal monitoring	438	37.2
3. Repair of current obstetric laceration	372	31.4
4. Circumcision of infant	321	27.0
5. Respiratory intubation and mechanical ventilation	319	20.4
6. Diagnostic cardiac catheterization, coronary arteriography	277	7.2
7. Cesarean section	264	32.4
8. Artificial rupture of membranes to assist delivery	261	34.2
9. Episiotomy	252	28.5
10. Blood transfusion	197	15.2

* The term "All-listed procedures" refers to all procedures performed during a hospital stay; patients often receive more than one procedure.

What Are the Most Common Procedures for Hospital Stays Billed to Private Insurance?

• Seven of the top 10 procedures billed to private insurers are related to pregnancy, childbirth, and newborn infant care. Private payers are billed for over half of these types of procedures.
• Private insurers are billed for just over a third of hospital stays in which diagnostic cardiac catheterization is performed, indicating that government (i.e., Medicare) bears the largest burden of paying for this most common procedure.
• Private insurers are billed for 7 in 10 hospital stays for hysterectomy.

Table 7. Most Common Procedures for Hospital Stays Billed to Private Insurance

All Listed Procedures*	Total number of discharges with this procedure category (in thousands)	Private insurer's share of hospital stays with this procedure (in percent)
1. Diagnostic cardiac catheterization, coronary arteriography	1,409	36.4
2. Procedures to assist delivery	1,130	55.0
3. Circumcision of infant	771	65.0
4. Repair of current obstetric laceration	721	60.9
5. Fetal monitoring	639	54.2
6. Episiotomy	566	64.0
7. Cesarean section	496	60.9
8. Artificial rupture of membranes to assist delivery	434	57.0
9. Hysterectomy	431	71.9
10. Respiratory intubation and mechanical ventilation	423	27.1

* The term "All-listed procedures" refers to all procedures performed during a hospital stay; patients often receive more than one procedure.

What Are the Most Common Procedures for Uninsured Hospital Stays?

• Among uninsured patients, 4 of the top 10 procedures are related to pregnancy, childbirth, and newborn infant care: procedures to assist delivery, fetal monitoring, repair of current obstetric laceration, and circumcision.
• Seventeen percent of hospital stays for alcohol and drug detoxification are not covered by insurance.

Table 8. Most Common Procedures for Uninsured Hospital Stays

All Listed Procedures*	Total number of discharges with this procedure category (in thousands)	Percent of hospital stays with this procedure that are uninsured
1. Diagnostic cardiac catheterization, coronary arteriography	123	3.2
2. Procedures to assist delivery	83	4.1
3. Respiratory intubation and mechanical ventilation	73	4.7
4. Alcohol and drug rehabilitation/detoxification	63	17.0
5. Fetal monitoring	54	4.5
6. Repair of current obstetric laceration	45	3.8
7. Circumcision of infant	45	3.8
8. Upper gastrointestinal endoscopy, biopsy	44	3.7
9. Blood transfusion	41	3.2
10. Computerized axial tomography (CT) scan, head	37	6.4

* The term "All-listed procedures" refers to all procedures performed during a hospital stay; patients often receive more than one procedure.

Return to Contents

Procedures in High-Volume Hospitals

Do Patients in Hospitals that Do Higher Numbers of Certain Procedures Have Lower Mortality Rates?

• Recent research* suggests that the outcomes, or results, of certain procedures are related to how many of them are performed in a hospital. These procedures often require high-technology support; but the exact nature of this "volume-to-outcome" relationship is not well understood.
• Hospitals that perform more than a specific number of certain procedures have fewer in-hospital deaths for those particular procedures. This "threshold number" varies with the type of procedure.
• For example, hospitals where at least 500 CABG procedures are performed each year have significantly lower mortality rates for CABG than hospitals performing fewer than 500. Hospitals doing nine or more heart transplantations annually have fewer in-hospital deaths for this procedure than hospitals doing fewer than nine.

---

* Adapted from Dudley R, Johansen K, Brand R, et al. Selective referral to high-volume hospitals: Estimating potentially avoidable deaths. JAMA 2000 Mar 1;283(9):1159-66.

---

Table 9. Ten Procedures with Evidence of a Volume-to-Outcome Link

Ten procedures with evidence of a volume-to-outcome link	Procedure-specific volume threshold: A low-volume hospital performs fewer than:
Elective abdominal aortic aneurysm repair	32
Carotid endarterectomy	101
Lower extremity arterial bypass	20
Coronary artery bypass graft	500
Coronary angioplasty	400
Heart transplantation	9
Pediatric heart surgery	100
Pancreatic cancer surgery	7
Esophageal cancer surgery	7
Cerebral aneurysm surgery	30

What Share of the Nation's Hospitals are High-Volume Providers of Specific Procedures?

• About 60 percent of hospitals that do heart transplantations, 55 percent of those that do lower extremity arterial bypass, and half of those that do coronary angioplasty are high-volume providers.
• Only about one-third of hospitals that perform coronary artery bypass graft surgery are high-volume providers.
• Fewer than 1 in 5 hospitals doing carotid endarterectomy and 1 in 6 performing pediatric heart surgery and abdominal aortic aneurysm repair are high-volume hospitals.
• Roughly 1 in 10 hospitals performing pancreatic cancer surgery, 1 in 15 doing cerebral aneurysm surgery, and 1 in 20 doing esophageal cancer surgery are high-volume providers.

Select Figure 6 (10 KB), What Share of the Nation's Hospitals are High-Volume Providers for Specific Procedures?

What Share of Patients Receive Procedures in High-Volume Hospitals?

• Over 90 percent of patients receiving lower extremity arterial bypass and about 85 percent of patients undergoing coronary angioplasty and heart transplantations receive them in high-volume hospitals.
• Just over 70 percent of heart surgery procedures for children are performed in high-volume hospitals.
• Less than two-thirds of coronary artery bypass graft surgeries are done in high-volume hospitals.
• Over half of all abdominal aortic aneurysm repair, carotid endarterectomy, and pancreatic cancer surgeries are done in high-volume hospitals.
• Only a fourth of surgical procedures for esophageal cancer and just over a third of those for cerebral aneurysm repair are done in high-volume hospitals.

Select Figure 7 (10 KB), What Share of Patients Receive Procedures in High-Volume Hospitals?

Return to Contents

Source of Data for this Report

The data presented in this report are drawn from the Healthcare Cost and Utilization Project (HCUP), a Federal-State-industry partnership to build a multi-State health care data system. This partnership is sponsored by the Agency for Healthcare Research and Quality and is managed by staff in AHRQ's Center for Organization and Delivery Studies. HCUP is based on data collected by individual States and forwarded to AHRQ by the States. HCUP would not be possible without State data collection projects and their partnership with AHRQ.

For 1997, 22 State organizations contributed their data to AHRQ where all data are edited and transformed into a uniform format. The uniform data in HCUP databases make possible comparative studies of health care services and the use and cost of hospital care, including:

• The effects of market forces on hospitals and the care they provide.
• Variations in medical practice.
• The effectiveness of medical technology and treatments.
• Use of services by special populations.

HCUP includes short-term, non-Federal, community hospitals (general and specialty hospitals such as pediatric, obstetrics-gynecology, short-term rehabilitation, and oncology hospitals are included). Long-term care and psychiatric hospitals are excluded.

HCUP includes two sets of inpatient databases for health services research. The State Inpatient Databases (SID) for 1997 cover inpatient care in community hospitals in 22 States and include nearly 60 percent of all hospital discharges in the United States. The Nationwide Inpatient Sample includes all discharges from a sample of about 1,000 hospitals drawn from the SID, adjusted to approximate a national sample.

This report is based on data from the NIS. The NIS approximates a 20-percent sample of U.S. community hospitals, as defined by the American Hospital Association (AHA). The NIS for 1997 includes information from 7.1 million discharges that were weighted to obtain estimates that represent the total number of inpatient hospital discharges in the United States (35.4 million).

Return to Contents

Methods

The Clinical Classification Software (CCS), developed by AHRQ, has been used throughout this Fact Book to aggregate procedure codes into a limited number of categories. Procedures recorded on hospital discharge records are coded using the International Classification of Diseases, 9th Revision, Clinical Modification (ICD-9-CM), Fifth Edition. Although ICD-9-CM may be used to provide descriptive statistics, aggregating similar diagnoses or procedures into clinically meaningful categories, such as the CCS, can be more helpful. Select for more information.

Frequencies and rankings of procedures are based on all-listed procedures; that is, all procedures listed on the discharge record. The unit of analysis is the inpatient stay, rather than the patient or procedure. All discharges have been weighted to produce national estimates.

Total charges in HCUP data are the amount the hospital charged or billed for the entire hospital stay and do not reflect charges for the individual procedures. Charges do not necessarily reflect reimbursements or costs and are generally higher than costs. Hospital charges do include professional (physician) fees. Charge data were present for 98 percent of all discharges.

Because the NIS is limited to inpatient hospital data, procedures performed in outpatient or ambulatory care settings are not reflected here.

Many medical terms are used throughout this report. For help in understanding these terms, refer to the Glossary.

.Return to Contents

Acknowledgments

Thanks to Suzanne Worth at Social and Scientific Systems for her invaluable assistance in statistical programming and to DonnaRae Castillo, AHRQ Office of Health Care Information, for editorial assistance.

Return to Contents
Proceed to Next Section