Chapter 9

Scope of Hospital Services:External Standards

and Guidelines

CONTENTS

Page

introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ......................187External Standards and Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. ...187

Standards for Overall Hospital Accreditation/Certification . ................187Standards and Guidelines for Specific Services . ...........................190

Reliability of the Indicator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..............197Validity of the Indicator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ...............197

Validity of Overall Hospital Accreditation . ..............................199Validity of Standards and Guidelines for Specific Services. . ................200

Feasibility of Using the Indicator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ...200Conclusions and Policy Implications. . . . . . . . . . . . . . . . . . . . . . .................202

BoxBox Page

9-A. Selected Sources of Information About Scope of Hospital Services ... ... ..2O3

TablesTable Page

9-2. Various Organizations’ Standards and Guidelines for Staffing NeonatalCare Facilities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ...............192

9-2. Characteristics of Various Organizations’ Standards and Guidelinesfor Emergency Services . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .........194

9-3. Specialty Organizations To Be Consulted in Developing the AmericanMedical Association’s “Guidelines for Classification ofHospital Emergency Capabilities,” January 1988..... . ...................195

9-4. HCFA’s Condition of Participation Governing Emergency Services . .......1959-s. Characteristics of Trauma Center Designations by State . ................1989-6. States That Require Copies of JCAHO Accreditation Reports

From Hospitals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .20~9-7. Characteristics of External Standards and Guidelines for Hospitals:

Overall Accreditation/Certification and Specific Services. . .........,.....204

Chapter 9

Scope of Hospital Services:External Standards and Guidelines

INTRODUCTION

Scope of hospital services is a structural meas-ure that reflects whether a hospital has the re-sources—facilities, staff, and equipment—to pro-vide care for the medical conditions it professesto treat or to care for the medical conditions affect-ing potential patients. There are several potentialsources of information on the scope of a hospi-tal’s services, including hospital advertising, me-dia reports about the existence of special equip-ment or specially trained staff, consumer guidelinesfor selecting medical providers, and organizationsthat accredit or certify hospitals.1 Identifyingwhether a hospital complies with external stand-ards such as those used for accreditation or cer-tification by an external body, however, is likelyto be the most valid means of ascertaining a hos-pital’s scope of services. Accreditations and cer-

IHospital certification typically refers to approval by governmentalbodies; accreditation usually indicates approval by a private orga-nization, most often a professional organization of peers. The term“guidelines” refers to standards proposed by professional organi-zations and voluntarily applied by providers.

tifications for scope of hospital services are dis-tinct from some of the other indicators evaluatedin this report in at least one sense. As currentlyconstructed, they measure only the capability ofa hospital to deliver good quality care, not thequality of care actually delivered or its outcome.

This chapter briefly describes two nationalmethods of overall accreditation/certification ofhospitals, that of the Joint Commission on the Ac-creditation of Heakhcare Organizations (JCAHO)and that of the Health Care Financing Adminis-tration (HCFA). It then describes external stand-ards and guidelines for neonatal intensive careunits, cancer care, and hospital-based emergencyand trauma services. The next sections of thechapter analyze the reliability, validity, and fea-sibility of using external standards and guidelinesrelated to the scope of hospital services as indi-cators of the potential of a hospital to deliver goodquality care. The final section draws conclusionsand discusses policy implications.

EXTERNAL STANDARDS AND GUIDELINES

Standards for Overall HospitalAccreditation/Certif ication

JCAHO Accreditation

The most well-known and widely applied hos- creditation, along with certain additional criteria,pital accreditation standards are those of JCAHO. is a condition of participation in the Medicare andOf the approximately 6,800 hospitals of all types Medicaid programs (Section 1865 of the Socialin the United States, about 5,000 (70 percent) are Security Act).2 Medicare and Medicaid pay forsurveyed by JCAHO. Submitting to JCAHOevaluation is voluntary, but not all hospitals are ‘In addition to being accredited by JCAHO, hospitals must meet

eligible for JCAHO surveys (325). One reason that requirements for utilization review (Section 1861(e)(6) of the So-

JCAHO accreditation is important is that such ac-cial Security Act (42 CFR Subpart S, 405.1901(d)(l) and 482.30) )and discharge planning (Public Law 99-190). In practice, the require-

(continued on next page)

187

188

about 38 percent of the hospital care provided inthis country (715). JCAHO accreditation is alsowoven through the hospital licensure requirementsof 41 States (323) and is a condition of participa-tion for an unknown number of insurance com-panies (48).

JCAHO conducts a complete survey of eacheligible hospital once every 3 years and assesseseach hospital’s compliance with over 2,000 stand-ards. The purpose of the JCAHO hospital accred-itation process is to evaluate each hospital’s over-all capability of providing medical care. Thus,particular attention is paid to functions affectingthe entire hospital, such as the governing body,the medical staff, nursing services, infection con-trol, and quality assurance, and the way these andother functions are integrated across the hospi-tal. Throughout this chapter, and for purposes ofevaluating JCAHO accreditation as a potential in-dicator of the quality of care, it is important tokeep in mind that it is not JCAHO’S purpose toseparately accredit individual hospital depart-ments such as those that provide emergency serv-ices or neonatal intensive care. Because JCAHOdoes survey and evaluate those services as partof its overall accreditation process, however,JCAHO standards for these separate departmentsare discussed in this chapter as having the poten-tial to evaluate whether hospital scope of serv-ices is appropriate.

JCAHO standards are developed by panels ofexperts, sometimes with the aid of scientific liter-ature, and are evaluated by interested hospitalsand other experts before their adoption. JCAHOstandards and required characteristics focus on

(continued from previous page)

ments for utilization review are met by the existence of utilizationand quality control peer review organizations.

In general, to meet the Medicare and Medicaid conditions of par-ticipation, hospitals must meet “any requirement under sectionM61(e) of the [Social Security] Act and implementing regulationswhich the Secretary [of Health and Human Services], after consultingwith [the Joint Commission] and [the American Osteopathic Asso-ciation], identifies as being higher or more precise than the require-ments for accreditation (section 1865(a)(4) of the Act)” (42 CFR Sub-part S, 405.1901(d)(3)). Psychiatric hospitals must meet “theadditional special staffing requirements that are considered neces-sary for the provision of active treatment in psychiatric hospitals(section 1861(f) of the Act) and implementing regulations” (42 CFRSubpart S, 405.1901(d)(2)).

certain key functions across the hospital: qualityassurance, privilege delineation, existence of pol-icies and procedures, and infection control.

A hospital’s failure to comply with key JCAHOstandards sometimes results in “accreditation withcontingencies. ” JCAHO gives each hospital a con-tingency score (which may be zero) that deter-mines in part whether the hospital is accredited.The actual accreditation decision is made byJCAHO’S Accreditation Committee, following arecommendation by JCAHO staff, using a set ofweighting procedures and objective rules to en-sure consistency across hospitals. If a hospital re-ceives a contingency, it must satisfy JCAHOwithin a specified period of time that it is in com-pliance with the problem standards. Dependingon the nature of the contingencies, hospitals mayhave to submit to a focused resurvey, usuallywithin 6 to 9 months from the date they receivethe report. From 1982, when the current JCAHOaccreditation procedure was implemented, until1987, the percentage of surveyed hospitals withJCAHO contingencies of any type increased fromabout 65 percent to 90 percent (387,388,524).

About 7S hospitals (S percent of JCAHO-sur-veyed hospitals) each year receive enough con-tingencies of a serious nature that a formal nonac-creditation decision from JCAHO looks probable;the hospitals are informed of this possibility byJCAHO staff before the staff recommendationgoes to the JCAHO Accreditation Committee.Among the s percent, 3 to 4 percent of theJCAHO-surveyed hospitals correct their deficien-cies to the satisfaction of JCAHO and avoid a for-mal nonaccreditation decision. Each year, about1 to 2 percent of all JCAHO hospitals surveyed,or 15 to 30 hospitals, are formally judged byJCAHO to be nonaccredited. Some of the 1 to 2percent of hospitals that are formally nonac-credited work on correcting deficiencies while theyare appealing the JCAHO decision and then re-quest a resurvey; others drop their quest forJCAHO accreditation, sometimes permanently.Some hospitals do, however, request HCFA in-spection following nonaccreditation by JCAHO.

HCFA Certification

Hospitals that desire Medicare and Medicaidreimbursement but choose not to be surveyed by

189

Photo credit: Joint Commission on the Accreditation of Healthcare Organizations

A JCAHO surveyor examines hospital records. JCAHO’S accreditation process is intended to evaluate the overall capa-bility of a hospital to provide medical care, rather than to evaluate particular services.

JCAHO or cannot meet JCAHO’S eligibility or ac-creditation criteria may opt to be certified byHCFA. About 1,400 hospitals per year routinelychoose to be surveyed by HCFA. Because torevery day that a hospital is not certified by HCFA,it loses Medicare and Medicaid reimbursement,not being accredited by JCAHO or certified byHCFA is very costly for a hospital.3

Most HCFA-certified hospitals are small, ru-ral community hospitals (438). Texas has thelargest number of HCFA-certified hospitals (1s7hospitals), followed by Kansas (83), Minnesota(63), Georgia (59), Nebraska (56), Mississippi (55),California (53), Oklahoma (51), Louisiana (50),Florida (48), and Iowa (46) (438). Those 11 Stateshave half the non-JCAHO-accredited, HCFA-cer-tified hospitals in the United States and its pos-sessions.

3Accreditation by the American Osteopathic Association enjoysthe same status with respect to Medicare and Medicaid paymentas JCAHO accreditation.

HCFA uses survey methods that are somewhatdifferent from JCAHO’S. HCFA’S hospital surveysare conducted annually, whereas JCAHO’S areconducted every 3 years. HCFA/State surveyorshave the force of law and the threat of noncer-tification to ensure compliance, while the JCAHOorganization does not. HCFA surveyors are Statepersonnel, and although the teams receive sometraining from HCFA, their composition is deter-mined by the States (399). JCAHO surveyors arehired and trained by JCAHO. JCAHO provides2 weeks of didactic training, a 3- to 4-week precep-torship, and an annual 3-day conference for sur-veyors.

JCAHO has stricter criteria for surveyors thandoes HCFA. JCAHO requires each survey teamto include one physician, one nurse, and one hos-pital administrator. In addition, JCAHO requiresthe nurse and hospital administrator surveyors tohave had administrative experience in a hospital.The qualifications of HCFA/State surveyors aremore diverse, and many of these surveyors arenot as highly trained as JCAHO surveyors. Of

IW

the 2,786 surveyors (of a total of about 3,400) whoresponded to a HCFA questionnaire, for exam-ple, only 10 (less than one-half of 1 percent) weremedical doctors (646). Finally, HCFA has substan-tially fewer standards than does JCAHO, andHCFA’S conditions of participation are much lessdetailed than JCAHO’S standards. Generally, 1percent or Iess of the hospitals surveyed by HCFAeach year are terminated from the program in-voluntarily (249).4

Overall Hospital Accreditation/Certificationand Scope of Services

Neither JCAHO accreditation nor HCFA cer-tification is designed to assess whether particularhospital departments are capable of providing spe-cific services. Nevertheless, JCAHO accreditationor HCFA certification does ensure that a certainscope of services exists in a hospital. In order toqualify for the survey on which JCAHO accredi-tation is based, a hospital must meet certain eligi-bility criteria. The hospital must maintain facil-ities, beds, and services that are available overa continuous 24-hour period, 7 days a week. Un-less a hospital is a psychiatric or substance abusefacility, it must also provide diagnostic radiology,dietetic, emergency, rehabilitation, and respira-tory care services, among others. In addition, itmust provide at least one of the following acute-care clinical services: medical, obstetric-gynecol-ogical, pediatric, surgical, psychiatric, or alcohol-or drug-abuse services. If the hospital providesobstetric-gynecological or surgical services, itmust also provide anesthesia services.

A hospital is also required to supply far fewerhospital services for HCFA certification than forJCAHO accreditation. Services required byJCAHO that are not required by HCFA includeemergency services, nuclear medicine services,some type of special care services, professionallibrary services, and social work services. For bothJCAHO accreditation and HCFA certification,surgical services are optional.5 Although both

41n fiscal year 1987, 9 hospitals were terminated involuntarily,in fiscal year 1986, 20 hospitals were terminated involuntarily, andin fiscal year 1985, 8 hospitals were terminated involuntarily fornot meeting HCFA’S conditions of participation.

‘The reason surgical services are optional for JCAHO is to makeit possible for psychiatric hospitals to be accredited. In most other

HCFA and JCAHO rate a number of specific de-partments or services (e.g., diagnostic radiologicservices, outpatient services, surgical and anes-thesia services), for the most part, neither ratescondition-specific services such as heart diseaseor cancer services.

Standards and Guidelinesfor Specific Services

Neonatal Intensive Care Services

In 1976, in the face of a proliferation of neonatalintensive care units, the Committee on PerinatalHealth’ proposed guidelines for the regionaliza-tion of U.S. maternal and perinatal health serv-ices (142). Underlying the concept of regionali-zation of these services is the idea that high-riskmothers and infants will be screened and referredor transported to the appropriate level of care.The Committee on Perinatal Health proposedthree levels of hospital care for perinatal services.Hospitals that served as regional centers and pro-vided the most sophisticated neonatal intensivecare were to be designated Level III facilities. Hos-pitals that provided neonatal intensive care butlacked some services provided in Level 111 facil-ities were to be called Level 11 facilities; and hos-pitals that provided normal newborn care withno special units for the care of seriously ill infantswere to be called Level I facilities.

In 1983, the American Academy of Pediatricsand the American College of Obstetricians andGynecologists more fully explicated the respon-sibilities and requirements of the three levels ofhospitals in the regional system of maternal andperinatal services. A document issued by theseorganizations specified guidelines for minimumnumber of beds, square footage per bed, person-nel, hospital structure, equipment, ancillary sup-port, and educational services for parents (15).

A recent analysis by OTA concluded that ne-onatal intensive care has been in large part respon-sible for the remarkable decline in U.S. neonatal

respects, psychiatric hospitals are held to the same standards as allother accredited hospitals.

‘The Committee on Perinatal Health was a joint effort by theAmerican Medical Association, the American College of Obstetri-cians and Gynecologists, the American Academy of Family Physi-cians, and the American Academy of Pediatrics.

191

mortality rates over the past 25 years and has con-tributed to improved long-term developmentaloutcomes for premature infants; the improvedsurvival of premature infants has not been accom-panied by an increase in the proportion of babieswith serious long-term disability (194).7 Accord-ing to OTA’s analysis, however, “an extremelypremature baby’s chances for survival and nor-mal development are in large part determined bywhere the baby is born” (194). The evidencestrongly suggests that the likelihood of survivalamong very low birthweight babies (babies weigh-ing under 1,500 grams at birth) is highest if thebaby is born in a hospital designated a Level IIIneonatal facility. When considering these conclu-sions, however, one should keep in mind that theyare based on some studies that were not method-ologically rigorous (i.e., studies that did not userandom assignment of newborns to compare Level1, II, or 111 facilities). Some studies have foundthat very low birthweight infants in Level III unitshad lower mortality rates than those in Level IIunits.

The concept of regionalization for perinatalservices has not been so well accepted by hospi-tals and physicians, however (194). Despite theexisting guidelines, there is no standard nationalapplication of what constitutes Level II or Level111 perinatal care (106). Ohio and some otherStates use the American Academy of Pediatrics/American College of Obstetricians and Gynecol-ogists guidelines to evaluate each hospital’s perina-tal services and assign levels accordingly (73,106).In California and most other States, however, theregional system of perinatal services is informal,and each hospital classifies its own services (344).

JCAHO applies standards for neonatal inten-sive care units in its overall hospital accreditationprocess (325), but these JCAHO standards aremuch less detailed and specific than the guidelinesof the American Academy of Pediatrics andAmerican College of Obstetricians and Gynecol-ogists. Table 9-1 illustrates some of the differencesbetween them in terms of staffing. JCAHO’Sstandards do not differentiate between Level 11 and

‘OTA did find, however, that there has been an increase in theabsolute number of survivors with serious long-term disability (194).

III neonatal intensive care. Even though JCAHOevaluates neonatal or other specific services aspart of its overall hospital accreditation process,consumers may want- to gocreditation to approvals bycialty organization.

Cancer Care

beyond JCAHO ac-the appropriate spe-

Being stricken with cancer creates great fearamong patients, and patients with cancer are in-tensely interested in finding the appropriate placefor treatment. At least three organizations of in-dependent observers have devised systems of ap-proval for cancer treatment centers:

● the American College of Surgeons,. the Association of Community Cancer

Centers, and● the National Cancer Institute.

There are substantial differences among them.

Cancer program approval by the AmericanCollege of Surgeons is granted following an ap-plication and a survey by three members of theCommission on Cancer. The four basic require-ments for American College of Surgeons approvalare as follows:

1.

2.

3.

4.

the existence of an established multidiscipli-nary cancer committee that meets quarterlyand provides the overall leadership of thecancer program;an established tumor registry with 2 yearsof patient data and 1 year of successful (min-imum 90 percent) patient followup;patient-oriented, multidisciplinary cancerconferences conducted weekly or monthly;andcompletion of two patient care evaluationstudies each year (27,28).

Failure to comply with any one of these require-ments results in either a l-year approval (versusthe usual 3-year approval) or, if there are othersignificant deficiencies, nonapproval. When a hos-pital first applies for approval, approval is notgranted if there are any deficiencies. The Amer-ican College of Surgeons has approved about1,200 cancer programs (356), and an additional400 to 500 are awaiting approval (469a). Centersapproved by the American College of Surgeons’

192

Table 9-1 .—Various Organizations’ Standards and Guidelines for Staffing Neonatal Care Facilities

JCAHO Standards for Staffing Neonatal Intensive Care Units

S. P.7.4.2. The director or other qualified physician designee in charge of the unit has at least 1 year ofrecognized special training and experience, as well as demonstrated competence, in neonatology.

S. P.7.4.3. Pediatric surgery is provided in the hospital, as required.S. P.7.4.4. Nursing care is supervised by a registered nurse who has training, experience, anddocumented current competence in the nursing care of high-risk infants.S. P.7.4.5. The nursing staff is proficient in teaching parents how to care for their infants at home.S. P.7.4.9. Radiologic technologists are familiar with X-ray techniques to be used with newborn infantsso that repetitive exposures are not necessaw.

American Academy of Pediatrics and American College of Obstetricians and Gynecologists’ Guidelinesfor Staffing Level 1, Level H, and Level Ill Neonatal Facilities

Level / Level //Chief of serviceOne physician responsible for perinatal Personnel

care (or codirectors from obstetrics Joint Planning:and pediatrics) Ob: Board-certified obstetrician with

certification, special interest,experience, or training in maternal-fetalmedicine;

Peals: Board-certified pediatrician withcertification, special interest,experience or training in neonatalogy

Other physicians:Physician (or certified nurse-midwife) at

all deliveriesAnesthesia servicesPhysician care for neonates

Supervisory nurseRegistered nurse in charge of perinatal

facilities

Staff nurselpatient ratioNormal labor 1:2Delivery in second stage 1:1Oxytocin inductions 1:2Cesarean delivery 2:1Normal delivery 1 :6-8

Level I plus:Board-certified director of anesthesia

servicesMedical, surgical, radiology, pathology

consultation

Ob: RN with education and experience innormal and high-risk pregnancy onlyresponsible

Peals: RN with education and experiencein treatment of sick neonates onlyresponsible

Level I plus:Complicated labor/delivery 1:1Intermediate nursery 1 :3-4

Level Ill

Codirectors:Ob: Full-time board-certified obstetrician

with special competence in maternal-fetal medicine.

Peals: Full-time board-certifiedpediatrician with special competencein neonatal medicine

Levels I and II plus:Anesthesiologists with special training or

experience in perinatal and pediatricanesthesia

Obstetric and pediatric subspecialists

Supervisor of perinatal sewices withadvanced skills

Separate head nurses for maternal, fetal,and neonatal services

Levels I and II plus:Intensive neonatal care 1:1-2Critical care of unstable neonate 2:1

Other personnelLicensed practical nurse, assistants

under direction of head nurseLevel I plus: Level I plus:Social sewice, biomedical, respirator Designated and often full-time social

therapy, laboratory as needed service, respiratory therapy, biomedicalengineering, Iaboratow technician

Nurse clinician and specialistsNurse program and education

coordinators

SOURCES: JCAHO standards: Joint Commission on the Accreditation of Healthcare Organizations, AmH/M: Accreditat)orr Manual for Hospitals (Chicago, IL: 19SS);AAP/ACOG guldelinas: American Academy of Pediatrics and American College of Obstetricians and Gynecologists, Guidelines for Perhrata/ Care (Evanston,IL: 19S3).

Commission on Cancer are listed in the American The American College of Surgeons’ patient careHospital Association’s Guide to the Health Care evaluation studies are similar to JCAHO’S moni-Eie]d (29), and a list of approved programs is toring and evaluation requirements,8 except thatavailable from the American College of Surgeons.The American College of Surgeons does not dis-

‘A key aspect of hospital quality assurance activities required byclose how many programs have been refused ap- JCAHO, the monitoring and evaluation process includes identify-proval, except to say that the number is small. ing important aspects of care, identifying indicators related to these

193

established American College of Surgeons-approved programs are required to complete onestudy to measure process and one study to meas-ure outcome, and new programs may completetwo process studies each year until sufficient dataare available to participate in the outcome studies.The specifics of both JCAHO’S and the AmericanCollege of Surgeons’ monitoring/evaluation pro-grams are determined internally at the hospital.9

Unlike JCAHO, however, the American Collegeof Surgeons requires that the outcome study com-pare the hospital’s experience with national or re-gional results (27). The American College of Sur-geons does have a voluntary program of cancerpatient care evaluation, in which results are com-pared across hospitals.

In comparison to the American College of Sur-geons’ program, the accreditation program of theAssociation of Community Cancer Centers is justbeginning. Membership in the Association ofCommunity Cancer Centers is granted if a can-cer center has the following:

1. a multidisciplinary cancer program;2. supervision by a multidisciplinary cancer

committee, group, or team; and3. direct or indirect involvement with care for

cancer patients.

Membership is open to freestanding cancercenters, health maintenance organizations, phy-sician group practices, home health agencies,hospital-based cancer programs and individualproviders (45). The Association of CommunityCancer Centers has standards, but they operateprimarily as guidelines to be used as self-assess-ment tools by the association’s organizationalmembers (46). The Association of CommunityCancer Centers has about 30 hospital membersand plans to begin a survey process in the nearfuture (179).

The National Cancer Institute has several pro-grams to designate cancer centers: the Compre-

aspects of care, establishing thresholds for evaluation related to theindicators, collecting and organizing data, evaluating care whenthresholds are reached, taking actions to improve care, assessingthe effectiveness of the actions and documenting improvement, andcommunicating relevant information to the organizationwide qualityassurance program (326).

‘JCAHO recently modified its requirements to encourage the useof indicators from the clinical literature (326).

hensive Cancer Centers program, the Community

Clinical Oncology Program, and the CooperativeGroup Outreach Program. Such designations area requirement for receiving support grants and arebased primarily on research capability (118, 580,668).

Emergency and Trauma Services

Emergency and trauma services involve situa-tions in which life or death may be at stake, andare therefore of extreme importance to consumers.In addition, consumers seem more likely to choosean emergency department than other hospital de-partments, although they may consult their phy-sicians for advice or direction. 10

There are several sources of standards andguidelines for the scope of emergency services thatmay potentially be of use to consumers. JCAHO,HCFA, the American College of Emergency Phy-sicians and Emergency Nurses Association(ACEP/ENA), and the American Medical Asso-ciation (AMA) all have or are planning guidelinesfor emergency services (23,36,38,325,642).11 TheAmerican College of Surgeons has a set of guide-lines for trauma care (26). In addition, manyStates and other localities have requirements thathospitals must meet to provide emergency serv-ices and/or to be designated as trauma centers.

Here as in other sections of this chapter, thedistinction must be kept in mind between stand-ards and guidelines. Ordy the requirements foremergency services and trauma care of JCAHO,HCFA, and States and localities are required foraccreditation or certification by those organiza-tions and can strictly be considered standards.Specialty organizations provide guidelines foremergency services and trauma centers, but theiruse by hospitals is optional. The ACEP/ENAguidelines, for example, are a “statement of sug-gested capability . . . not designed to be inter-preted as mandatory by legislative, judicial, or

IOTheir phy&cianS may be on the medical staff of a particular hos-pital and may direct the patient to that hospital so they may carefor the patient there.

llThe Accreditation Association for Ambulatory Health Care alsohas standards for emergency services, but their standards are ori-ented primarily toward freestanding emergency service centers (4).The ACEP/ENA guidelines apply to both hospital and freestand-ing emergency facilities (23).

194

regulatory bodies” (23). Similarly, AMA guide-lines for emergency services, currently under re-vision, are to be considered guidelines for use byhospitals, rather than standards (36,38,178). Nei-ther ACEP/ENA nor the AMA has any plans tosurvey for compliance with the guidelines theyhave devised. ACEP/ENA, American College ofSurgeons, and AMA guidelines have, however,been adopted or adapted by some State bodiesfor regulatory use.

The following discussion makes a distinctionbetween emergency services and trauma centers,but in practice, the distinction is not always clear.The medical services being evaluated in thetrauma literature are not always restricted totrauma care (543), and there is some overlap inthe guidelines for emergency services and traumacenters, as there is in the services themselves.Some trauma centers have their own admittingareas and staff (621), while others are a “concept”within the emergency department (153). In gen-eral, however, emergency medical services focuson prehospital care and care within the emergencydepartment; trauma care includes inhospital andrehabilitative care.

Standards and Guidelines for Emergency Serv-ices.—Standards and guidelines for emergencyservices and standards can be distinguished alongat least three dimensions: 1) whether they arestandards or guidelines, 2) their breadth or depth,and 3) whether they distinguish among levels ofservices. Table 9-2 shows how various organiza-tions’ standards and guidelines for emergencyservices can be characterized along thesedimensions.

The proposed AMA guidelines for emergencyservices will have perhaps the largest breadth, be-cause they will be a compilation of guidelines fromabout 10 specialty organizations. The list of spe-cialty organizations consulted in the developmentof the proposed AMA guidelines is shown in ta-ble 9-3.

Because the AMA guidelines for emergencyservices will incorporate the standards of specialtyorganizations, they will also have the greatestdepth. The guidelines for emergency services ofspecialty organizations such as the ACEP/ENA,for example, designate administrative and man-agerial responsibilities, staffing levels, equipment,drugs, and relationships among the emergencyservice and other hospital departments (23). TheACEP/ENA guidelines do not specify guidelinesfor care of specific conditions such as burns orpoisonings. Although the ACEP/ENA guidelinesdo not require that emergency departments oper-ate continuously and do not stipulate levels ofemergency care, they state that the emergency de-partment should be staffed by a physician dur-ing all hours of operation. Optimally, accordingto the ACEP/ENA guidelines, the medical staffshould be board certified in emergency medicineand the nursing staff should practice in accord-ance with the Standards of Emergency NursingPractice.

Like the other organizations, JCAHO lists vari-ous aspects of hospital emergency services: orga-nization, direction and staffing; integration, train-ing and education, policies and procedures; andfacility design and equipment. JCAHO standardsfor emergency services are more specific than

Table 9“2.—Characteristics of Various Organizations’ Standards and Guidelines for Emergency Services

Organization Standards or guidelines Breadth v. de~th Levels of care

JCAHO . . . . . . . . . . . . . . . . . . . . Standards a Breadth Levels I (highest) to IV (lowest)HCFA . . . . . . . . . . . . . . . . . . . . . . Standards a Breadth None specified

ACEP/ENA . . . . . . . . . . . . . . . . . Guidelinesb Breadth None specified

AMA . . . . . . . . . . . . . . . . . . . . . . . Guidelines b Breadth and depth To be specifiedAbbreviations: ACEP/ENA = American College of Emergency Physicians and Emergency Nurses Association; AMA = American Medical Association; HCFA = Health

Care Financing Administration; JCAHO = Joint Commission for the Accreditation of Healthcare Organizations.aThese guidelines apply to hospitals onlY.bThese guidelines apply to freestanding emergency facilities as well aS hospitals.

SOURCE: Office of Technology Assessment, 19S8.

195

Table 9-3.—Specialty Organizations To Be Consultedin Developing the American Medical Association’s

“Guidelines for Classification of HospitalEmergency Capabilities,” January 1988

Organization providingType of emergency guidelines or guidance

General medical . . . . . . American College of EmergencyPhysic iansa

Behavioral andpsychiatric . . . . . . . . . American Psychiatric

AssociationBurn . . . . . . . . . . . . . . . . American Burn AssociationCardiac . . . . . . . . . . . . . . American College of Cardiology

and American HospitalAssociation

Pediatric . . . . . . . . . . . . . American Academy of PediatricsPerinatal . . . . . . . . . . . . . American College of Obstetrics

and Gynecology and AmericanAcademy of Pediatrics

Poisoning or drug . . . . . American Association of PoisonControl

Spinal cord . . . . . . . . . . American Spinal Cord InjuryAssociation

Trauma . . . . . . . . . . . . . . American College of SurgeonsPediatric trauma . . . . . . American Pediatric Surgery

Association; American Collegeof Surgeons

~entative,

SOURCE: P. Dietz, Program Administrator, Commission on Emergency MedicalServices, American Medical Association, Chicago, IL, personal corn.munication, Jan 28, 1988.

ACEP/ENA guidelines with respect to the com-ponents of medical records for emergency patientsand include requirements for quality control andmonitoring and evaluation. In addition, JCAHOhospital-wide standards (e.g., medical staff re-quirements) apply to emergency services. Unlikethe ACEP/ENA guidelines, JCAHO standards re-quire that a hospital’s emergency service be clas-sified according to four levels of services provided,ranging from a “comprehensive” level of care(Level I) to a “first aid/referral” level of care (LevelIV). The primary distinguishing feature among thefour levels of emergency services is physicianavailability, although there also are differenceswith respect to nursing staff and equipment.

HCFA’S condition of participation governingemergency services is rather broad (see table 9-4). They do, however, contain some of the samebasic requirements as do the standards and guide-lines of other groups. These requirements pertainto organization and direction and the qualifica-tions of personnel. HCFA does not require spe-cific staff coverage, equipment, or drugs.

Table 9-4.—HCFA’S Condition of ParticipationGoverning Emergency Semices

482.55 Emergency SewicesThe hospital must meet the emergency needs of patients inaccordance with acceptable standards of practice.

a. Standard: Organization and direction.If emergency services are provided at the hospital:1. The services must be organized under the direc-

tion of a qualified member of the medical staff;and

2. The services must be integrated with other depart-ments of the hospital.

3. The policies and procedures governing medicalcare provided in the emergency service or depart-ment are established by and are a continuingresponsibility of the medical staff.

b. Standard: Personnel.1. The emergency services must be supervised by a

qualified member of the medical staff.2. There must be adequate medical and nursing per-

sonnel qualified in emergency care to meet thewritten emergency procedures and needs antici-pated by the facility.

SOURCE: U.S. Department of Health and Human Services, Health Care Financ-ing Administration, “Appendix A: Interpretive Guidelines and SurveyProcedures—Hospitals,” State Operations Manual, Provider Certifica-tion, HCFA-Pub. 7 (Baltimore, MD: September 1988).

Standards and guidelines for emergency serv-ices differ in their requirements regarding physi-cian services. ACEP/ENA guidelines for emer-gency care recommend that emergency facilitiesbe staffed during all hours of operation by a phy-sician” trained and experienced in emergency medi-cine. According to ACEP/ENA, unless there isphysician staffing, a hospital should not beregarded as able to provide emergency services(709). This is a somewhat controversial recom-mendation. Not all of the 77 million visits to emer-gency facilities in a year (506) require a physiciantrained and certified in the specialty of emergency

medicine, or even a physician. The basis of thisACEP/ENA guideline, however, is that “emer-gency health care exists for the individual bene-fit of the patient or family who perceives a needfor emergency care, and for society’s benefit inmost casualty accidents” and that “the Americanpublic justifiably expects an emergency facility tobe staffed by medical, nursing, and ancillary per-sonnel who are trained and experienced in thetreatment of emergencies” (23).

JCAHO’S standards for emergency services donot require the presence of a physician at all times.JCAHO’S standard for Level IV, the least com-

Photo credit: American College of Emergency Physicians

External standards and guidelines for emergency serv-ices differ on whether a physician must be available

at all times in hospital emergency rooms.

prehensive level of care, for example, is that the“emergency service offers reasonable care in de-termining whether an emergency exists, renderslifesaving first aid, and makes appropriate refer-ral to the nearest facilities that are capable of pro-viding needed services. ” There must be somemechanism for providing physician coverage atall times in Level IV emergency facilities, but themechanism is to be defined by the medical staffof the hospital. That the standard does not requireimmediate availability is reflected in JCAHOstandards for Level III and higher emergency fa-cilities. Level III facilities, for example, are re-quired to have at least one physician available tothe emergency care area within approximately 30minutes. 12 The impact of having a trained and ex-perienced physician available in an emergency de-partment at all times has not been evaluated, sothe relative validity of these standards cannot bejudged.

In addition, it is noteworthy that only one setof standards or guidelines for emergency serv-ices—JCAHO’s—requires that a hospital have aprovision for providing emergency care 24 hoursa day, 7 days a week (325).

‘zLevel I and 11 hospitals are required to have at least one physi-cian experienced in emergency care on duty in the emergency carearea at all times. In addition, in Level I hospitals, there must be in-hospital physician coverage by members of the medical staff or bysenior-level residents for at least medical, surgical, orthopedic, ob-stetric/gynecological, pediatric, and anesthesiology services.

Trauma Center Designations.-A review by theCenters for Disease Control of mortality data for1984 shows that unintentional injuries were theleading cause of “years of potential life lost” be-fore the age of 65 (440). A large proportion of ef-forts to decrease the number of deaths caused byinjury have focused on injury prevention, but con-siderable attention has also been directed to thedesignation and implementation of emergencymedical service systems (e.g., 454, 455; the Fed-eral Emergency Medical Services Systems Act of1973 [revised in 1975, repealed in 198113]). In anorganized emergency system, some hospitals aredesignated as regional trauma centers, to whichseverely multiply injured individuals are broughtfor treatment.

Intuitively, one expects that treatment and out-come in trauma centers will be better than else-where because of the immediate availability ofrapid transportation, highly trained field person-nel and emergency physicians, modern diagnos-tic tools, and experienced trauma surgeons (543).The only current national guidelines for traumacenters have been devised by the American Col-lege of Surgeons’ Committee on Trauma (26). TheAmerican College of Surgeons’ guidelines incor-porate resources for both prehospital and hospi-tal care. For hospitals, the guidelines specify thedesired characteristics for three levels of traumacare. The two highest levels (Levels I and 11) havesimilar requirements for patient care; the highestlevel (Level I) has additional requirements for edu-cation and research in trauma. Level III traumacenter hospitals serve communities that do nothave alI the resources usually associated withLevel I or 11 institutions; Level 111 facilities musthave a “maximum commitment to trauma carecommensurate with resources. ” Thus, for exam-ple, a Level 111 hospital might have a surgeon andother personnel on call rather than in-house.Nonetheless, a Level 111 facility would be calleda trauma center by the American College ofSurgeons.

According to a recent survey by the AmericanCollege of Surgeons, approximately 177 hospitals

ls~e F~eral Government devolved much of its leadership respon-sibilities to States by folding the Emergency Medical Services Sys-tems Act program into the Preventive Health and Health Servicesblock grant.

197

have Level I trauma centers, 138 of which are des-ignated as Level I by some external body; the re-mainder are self-designations by hospitals them-selves. About 157 hospitals have Level II traumacenters, 124 of which are so designated by someexternal body (127). Table 9-5 indicates that only19 States designate trauma centers using either theguidelines of the American College of Surgeonsor a modified version of those guidelines.

The availability of various surgical, as opposedto medical, personnel is a major requirement formeeting the American College of Surgeons’ guide-lines, although there are numerous other require-

RELIABILITY OF THE INDICATOR

Accreditation schemes for hospitals overall andfor particular services are, it is clear, highly vari-able. To a consumer interested in neonatal inten-sive care, certification by the State of Ohio fora particular level of neonatal intensive care wouldconvey much more information than the fact thata hospital with a neonatal intensive care unit hadreceived JCAHO accreditation. Similarly, to aperson interested in cancer care, a hospital’s mem-bership in the Association of Community Can-cer Centers or designation as a ComprehensiveCancer Center by the National Cancer Institutewould not convey the same type of approval aswould approval by the Cancer Commission of theAmerican College of Surgeons. Overall, HCFA’Scertification process is not as rigorous as JCAHO’Saccreditation process.

Some States have developed specific require-ments for hospitals to offer specific services, but

ments as well (26). For example, the AmericanCollege of Surgeons recommends that a traumateam be organized and directed by a surgeon. Thesurgeon-directed trauma team is to evaluate thepatient initially, and a surgeon is to be responsi-ble for the patient’s overall care. A physician withspecial competence in care of the critically injuredis to be a designated “member” of the traumateam, and is to continuously staff the overallemergency department, but not be the head of thetrauma team. Although the need for surgeons todeliver most trauma care is generally acknowl-edged, there is some controversy about whoshould design and manage the overall service (44).

the types of services under these regulations andthe specific requirements differ across States. InCalifornia, for example, emergency services areconsidered a supplemental service and appear assuch on hospital licenses and published informa-tion for consumers (113,345); New York is aboutto change a similar regulation to make emergencyservices a basic requirement (472).

At the level of the individual State standard,there is considerable variation, because States de-velop their standards through statute and regu-lation, and statutes vary across States. The relia-bility of the surveyors and the survey process mayvary as well. Hospitals surveyed by JCAHO, forexample, have complained that judgments regard-ing their compliance with the same standard mayvary considerably between survey periods. Inpart, the variation is due to periodic revision byJCAHO of its standards, a necessity.

VALIDITY OF THE INDICATOR

Accreditation for scope of hospital services is that much of medical practice is not based on evi-not a single entity, and individual standards them- dence from scientific studies (628). Decisionsselves may vary in the extent to which they have about the “best” staff, equipment, and organiza-been validated. Optimally, perhaps, standards tion for a particular service or a particular prob-and guidelines for scope of services would be lem are often the result of clinical judgment. Thus,based on medical practice with systematically most standards have been developed through ex-demonstrated efficacy. The problem, however, is pert consensus.

198

c)n n

<>

<>

z

z

5

z

5

. .. .. .. .. .. ..,. .. .. .. .. .

aIzlji

ea

n

Inal51-

E< c)on u )-1

Jc

~

1-3

. . . . . . . . . . .. . . . . . . . . . .. . . . . . . . . . .. . . . . . . . . . .. . . . . . . . . . .. . . . . . . . . . .. . . . . . . . . . .. . . . . . . . . . .. . . . . . . . . . .. . . . . . . . . . .. . . . . . . . . . .

. . . . .. . . . .. . . . .. . . . .. . . . .. . . . .. . . . .. . . . .. . . . .. . . . .. . . . .

. . . .. . . .. . . .. . . .. . . .. . . .. . . .. . . .. . . .. . . .. . . .. . . .. . . .. . . .

. . . . . .. . . . . .. . . . . .. . . . . .. . . . . .. . . . . .

. .. . . . .. . .. . .. . .. . .. . .. . .. . .. . .. . .. . .. . .. . .. . .. . .. . .. . .. . .

. . . . . . . . .. . . . . . . . .. . . . . . . . .. . . . . . . . .. . . . . . . . .. . . . . . . . .

. . . .. . . .. . . .. . . .. . . .. . . .. . . .. . . .. . . .. . . .. . . .. . . .. . . .. . . .. . . .. . . .

. .. .. .. . . . . . . . . . .. . . . . . . . .. . . . . . . . .. . . . . . . . .. . . . . . . . .. . . . “w : ; ;. . . . .-. .. .. .. .. .. .

. . . .= . . . . . .

199

Expert consensus may be an appropriate basisfor establishing standards and guidelines for hos-pitals overall and for particular conditions, serv-ices, and departments. For some services, how-ever, groups of experts disagree with one another,either on the need to establish standards or on thecontent of the standards themselves. The con-sumer is then left with the puzzling question ofwhich group of standards or guidelines is morevalid.

The validity of particular standards and guide-lines could be demonstrated with studies of rela-tionships between standards or guidelines andgood process and outcomes, determined post hoc.Some standards and guidelines, such as those forneonatal intensive care and trauma centers, havebeen subjected to some such study, but most havenot. JCAHO’S hospital accreditation standards forhospitals have been subjected to very little study,and HCFA’S hospital certification standards sub-jected to none. The studies that have been con-ducted have had methodological problems. Forthe most part, they have relied on retrospectiveanalysis and outcomes as criteria and have notbeen conducted by independent observers.

One significant problem, applicable to all stand-ards and guidelines, is that the standards or guide-lines may change over time, sometimes signifi-cantly (388), a situation that makes the results ofstudies conducted at one point in time not appli-cable to subsequent standards. Frequent changesin standards may, of course, be necessary to re-flect changes in technology and medical practice.

Validity of Overall HospitalAccreditation

There has been little attempt to validate over-all JCAHO accreditation as an indicator of thequality of care. An important factor limitingstudies seeking to validate JCAHO accreditationis that accreditation is refused or withdrawn forso few hospitals that the mere fact of accredita-tion may not be very sensitive to variations inquality. The few studies of the validity of JCAHO

1dAt the time Hyman collected his data, JCAHO was using theterminology “recommendations” rather than “contingencies. ”

accreditation as an indicator of the quality of carehave yielded inconclusive or noncomparableresults.

Hyman obtained the results of JCAHO surveysfor New York City hospitals (312). Unexpectedly,Hyman found that publicly supported hospitalshad better JCAHO contingency scores14 thanvoluntary not-for-profit hospitals on 9 of 11 func-tions. Friedman analyzed the relationships be-tween numbers of JCAHO contingencies andHCFA’S 1984 hospital mortality data (237). Theresult was a very low, statistically insignificantcorrelation, but this result is not surprising giventhe problems with HCFA’S measure of hospitalmortality (see ch. 4). One internal JCAHO studyfound a high level of agreement among JCAHOsenior clinical and administrative staff as to thesignificance of several categories of standards forensuring quality patient outcomes, but actual out-comes or process criteria were not used as vali-dation standards (572).

Because JCAHO accreditation means thathospitals will be certified by HCFA, HCFA is re-quired by law to validate JCAHO’S results (Sub-section 1864(c) of the Social Security Act). Ev-ery year, HCFA requests that State surveyorssurvey a small sample of JCAHO-accreditedhospitals, stratified to be representative of hospi-tals nationally. HCFA also asks State surveyorsto investigate patient complaints that seem to havesubstance. The State surveyors perform JCAHOvalidation surveys for HCFA using the Medicareconditions of participation. If a State surveyorfinds that a hospital has significant deficienciesthat could affect the health and safety of patients,the hospital is placed under State surveillance untilthe deficiencies are corrected. The hospital is nolonger deemed to meet the Medicare conditionsof participation, and the State monitors the cor-rection of any deficiency.

HCFA conducted the last published JCAHOvalidation survey in fiscal year 1983, and trans-mitted it to Congress in 1986 (639). In general,JCAHO hospitals were found to be in compliancewith HCFA’S requirements. Any conclusion thatJCAHO standards are valid because of their com-pliance with HCFA’S requirements, however, de-pends on the validity of HCFA’S survey process,

and that process has not been validated. In addi-tion, the discrepancy rates that HCFA found be-tween HCFA’S deficiencies and JCAHO’S contin-gencies would mean that 276 hospitals in anysingle year, and as many as 750 hospitals overallls

would be out of compliance on some conditionof participation.

One future source of information for develop-ing and validating JCAHO (and HCFA) standardsis JCAHO’S Agenda for Change project (see app.D). This project is attempting to develop morevalid and condition-specific standards, includingclinical process and outcome indicators. A poten-tial JCAHO clinical indicator for obstetrics, forexample, is birthweight-specific hospital mortal-ity rates; hospitals designating themselves as highlevel neonatal intensive care units may have tomeet a minimum birthweight-specific mortalityrate. This project is being pilot-tested now witha small sample (324). In addition, JCAHO isprogressing with plans to revamp its structural in-dicators so that they reflect the characteristics ofeffective health care organizations.

Validity of Standards and Guidelinesfor Specific Services

Many of the available studies of the validityof trauma center designations as indicators of thequality of care are methodologically flawed.Those that rely in whole or in part on autopsy

15 H(3A’s 1983 ValjdatjOn surveys found that Up to 15 percent ofhospitals surveyed were not in compliance with HCFA standards,although they had been in compliance with JCAHO’S standards.If the 15 percent noncompliance rate is multiplied by the total numberof JCAHO-surveyed hospitals (5,000), the number of hospitals notin compliance with HCFA standards would be 750.

studies, for example, are biased in that not alldeaths result in autopsies. Some studies use differ-ent sources of information to determine causes ofdeath. In one study of the San Diego CountyRegional Trauma System, for example, the causesof deaths in trauma centers were taken from atrauma registry, but the causes of death in com-parison hospitals were taken from autopsies (564).

Perhaps more important, most studies oftrauma center designations tend to be uncon-trolled; that is, they merely compare patient out-comes before and after implementation of atrauma system. Studies that merely compare out-comes before and after implementation of atrauma system do not take into account factorsother than medical care that may be responsiblefor reducing death rates from trauma (543). Thesefactors may include simultaneous changes, suchas reductions in speed limits and enhanced en-forcement of drunk driving laws. In studies ofstandards and guidelines for neonatal intensivecare, most of the research has been done only onLevel III neonatal intensive care units (194), andthe validation standards have been outcome meas-ures, primarily mortality. Plans are underway toconduct studies of neonatal intensive care unitsusing process criteria for validation.

Standards for emergency services have not beensubject to the same amount of study that traumacenter designations have, perhaps because thescope of services in emergency rooms is so broad.A knowledgeable observer concluded that there isno dependable knowledge about interhospitaldifferences in emergency department performanceor about the sources and correlates of such differ-ences; there is also no dependable knowledge aboutthe factors and conditions that facilitate or hin-der emergency department effectiveness (245, 246).

FEASIBILITY OF USING THE INDICATOR

If validated, compliance with external stand- post. JCAHO’S certificate addresses overall hos-ards for scope of hospital services is potentially pital accreditation, not individual services.an extremely valuable and easily accessible indi- Detailed reports on the results of JCAHO surveyscater of the quality of care for consumers. Cur- of hospitals would be more informative; but theserently, JCAHO and the American College of Sur- results are for the most part not easily obtained.geons both provide hospitals with a certificate to

201

JCAHO releases to the public, on request, in-formation about whether a hospital is accredited,is involved in an appeal of its accreditation, isnonaccredited, or holds no accreditation status.JCAHO also releases a hospital’s accreditation his-tory. It does not, however, reveal a hospital’s con-tingency score or copies of the survey reports. TheJCAHO survey reports may be available on re-quest from individual hospitals and from thoseStates that require hospitals to submit the detailedsurvey reports as a requirement for licensure.Some States make the survey reports available;New York, Pennsylvania, and Arizona are amongthem. Other States, including California and 11-linois, do not release copies of the JCAHO sur-vey reports. States that recognize JCAHO accred-itation for State hospital licensure purposes andrequire a copy of the accreditation report fromthe hospital are listed in table 9-6. JCAHO sur-vey reports are long and technical, and consumersmay face problems in interpreting the informa-tion they contain. One problem is that the sur-vey reports focus on what is wrong with the sur-veyed hospitals. Without reviewing survey reportsof several hospitals, consumers would not beaware of how a particular hospital compared withother hospitals.

Results of HCFA’S hospital surveys exist in sev-eral forms. HCFA constructs individual hospitalfacility profiles that indicate the types of deficien-cies a hospital has had for past survey years, andthe services and personnel available at the hospi-tal, among other information (649). In addition,HCFA constructs a table comparing State, re-gional, and national deficiency patterns for eachMedicare condition of participation (650). A ta-ble constructed in January showed thats (27 per-cent) of the 18 HCFA-inspected hospitals in oneState had deficiencies in the area of licensure ofpersonnel; this rate compares to 19 percent forthe U. S. Department of Health and Human Serv-ices Region 111 and 13 percent for the Nation (650).Some of the information from HCFA is not easyto use, however. The individual facility profilesreport deficiencies by code numbers. These codenumbers are not the same as the information onthe report for the State, region and Nation, whichdoes include written descriptions of the HCFAconditions of participation. It is, however, easy

to glean from the individual facility profiles theservices available at the hospital, which could bean important source of information for con-sumers. Both HCFA reports are intended as in-ternal management tools for HCFA, but must bemade available to the public on request (249). Asfor the individual survey reports, copies of a re-port (form 2567) that includes both the surveyors’recording of deficiencies and the hospital’s planof correction, and copies of the original surveyreports from which the deficiency portion of form2567 is drawn, are available from State surveyoffices, which are required to release them to thepublic (249).

Some States publish information about hospi-tal accreditation and certification overall, licen-sure for particular services, and other informa-tion. California, for example, will send consumerswho request it a summary report on hospitals.Hospitals in California and New York State mustpost in a conspicuous place their licenses, whichnote the services that the hospital is permitted toprovide (34s,414).

The feasibility of using scope of service desig-nations to indicate quality of care is affected bythe tendency of hospitals to self-designate them-selves as specialists in particular areas. Even someState approval of trauma centers is based on hos-pital self-designation. Consumers would have tobe careful that a designation is based at least onthe stipulation of independent observers that thehospital adheres to a set of standards; otherwisesuch a designation may not be a valid indicatorof quality. The American Hospital Association

Table 9-6.—States That Require Copies ofJCAHO Accreditation Reports From Hospitals

ArizonaCaliforniaConnecticutDistrict of ColumbiaFloridaGeorgiaIdahoIllinoisIowaKansasLouisianaMaine

MassachusettsMinnesotaMississippiMontanaNebraskaNew HampshireNew YorkNorth CarolinaPennsylvaniaSouth CarolinaUtahWvomina

SOURCE: Joint Commission on the Accreditation of Healthcare Organizations,“State Project Status Report, ” Chicago, IL, Sept. 21, 19S7.

202

currently publishes a guide indicating the facil-ities and services available at hospitals that par-ticipate in the association’s survey, but thesedesignations are based largely on hospital self-reports. The American Trauma Society also pub-lishes a list of trauma centers based on self-designation.

Consumers also face the problem of conflict-ing sets of standards for the same service. For can-cer care, for example, there will soon be stand-ards from two organizations (the AmericanCollege of Surgeons and the Association of Com-munity Cancer Centers). Although these stand-

ards build on each other to some extent, their rela-tive validity remains to be established.

Even if available and reasonably validated,however, accreditation and standards for scopeof services rely on the ability of patients to“match” their condition with the service as de-scribed by the accrediting body. When a patientrequires more than one service, the problem be-comes even more complex. Even accreditationsthat seem relatively condition-specific may not beuseful to a particular patient. Hospitals whosecancer programs are approved, for example, maybe more successful with some types of cancer thanothers.

CONCLUSIONS AND POLICY IMPLICATIONS

The external standards and guidelines that havebeen promulgated for hospital services overall andfor scope of hospital services have not beenrigorously validated as indicators of quality ofcare. Clearly, however, it seems worthwhile forconsumers to seek out hospitals that have beenjudged by independent experts to have the appro-priate resources to provide care, either overall orfor specific conditions.

Some accreditation/certification information isreadily available to consumers (see box 9-A). In-formation on a hospital’s JCAHO accreditationhistory, for example, is available from JCAHO.HCFA will provide information on the certifica-tion status of any of the approximately 1,400 hos-pitals it inspects, and the American College of Sur-geons will provide a list of the cancer programsit has approved. HCFA-inspected hospitals’ ac-tual survey reports are available from State agen-cies that conduct the surveys on behalf of HCFA.Some States require that hospitals post a noticestating which services they are allowed to performand others provide consumers with reports sup-plying such information.

Other information is in existence but is moredifficult for consumers to obtain or interpret.JCAHO survey reports, which form the basis ofJCAHO accreditation decisions, are an example.Such reports can provide more detailed informa-tion to consumers than the mere fact of JCAHO

accreditation, To see the reports, consumers mayhave to approach the hospitals themselves and askfor the reports, although some States will provideconsumers copies of JCAHO survey reports forindividual hospitals. Some consumers may havetrouble interpreting and comparing detailed sur-vey reports, and may prefer to see summary judg-ments that compare hospitals along a range ofscores. Although JCAHO computes overall con-tingency scores for hospitals and also evaluateswhether hospital emergency services meet require-ments for four levels of care, this information isnot readily available to the public.

There are considerably more guidelines avail-able for the internal, optional use of hospitals thanthere are standards applied by independent groupsof observers. Although hospitals may diligentlyconform to such guidelines, consumers should bewary of hospitals that say they adhere to the prin-ciples of one group or another, when there is noindependent evaluation of such compliance.

Several steps could be taken to address the ex-isting problems of external standards for overallhospital accreditation and scope of hospital serv-ices as quality-of-care indicators, to make exist-ing information available, to improve existingstandards, and to develop new standards. Table9-7 shows the status of existing standards andguidelines in terms of their validity and feasibil-ity of use as indicators of quality.

203

Box 9-A.—Selected Sources of Information About Scope of Hospital Services

Type of information Organization, address, or telephone number

JCAHO hospital accreditation history Joint Commission on Accreditation ofHealthcare Organizations

1-800-621-8007 (nationwide except Illinois)1-800-572-8089 (Illinois)

JCAHO hospital survey reports Available from States of New York,Pennsylvania, and Arizona

May be available from individual hospitals

HCFA hospital survey reports Available from State agencies that conductsurveys on behalf of HCFA

List of hospital cancer programs approved by Cancer Department

the American College of Surgeons American College of Surgeons5!5 East ErieChicago, IL 60611

With some effort, existing information aboutcompliance with existing standards could be madeavailable to consumers. It seems ironic, for ex-ample, that survey reports for the 1,Q O O p r e d o m -i n a n t l y s m a l l a n d r u r a l h o s p i t a l s s u r v e y e d b y

HCFA are available to the public, while surveyreports for the 5,000 hospitals surveyed byJCAHO on HCFA’S behalf are not. Hospitals ac-credited by JCAHO are paid on the same basisas those certified more directly by HCFA. HCFAcould improve its individual facility profiles, sothat the reasons for deficiencies are intelligible toconsumers and comparable to the reasons inHCFA’S State, regional and national reports. Asanother example, JCAHO could include as partof its accreditation certificate the level of emer-gency services provided at a hospital, so that con-sumers could know whether a physician was likelyto be on site. JCAHO and HCFA could developsummaries of their hospital survey reports thatare meaningful to consumers (e.g., they could de-vise summary scores for specific services). Suchinformation could be made available at hospitalsthemselves and in public places, such as libraries,local government offices, Social Security offices,and the offices of utilization and quality controlpeer review organizations. Similar informationabout the approvals by professional specialtyorganizations could also be made available.

Research to validate existing standards and helpdevelop new standards is essential if consumersand providers are to be able to have confidencein the standards. Research is needed on all thestandards and guidelines for scope of hospitalservices discussed in this chapter: JCAHO hospi-tal accreditation standards; HCFA hospital cer-tification standards; and various organizations’standards and guidelines for neonatal intensivecare units, cancer care, emergency services andtrauma units. Undoubtedly, research is needed onother condition-specific services. As some of theorganizations that have developed standards be-gin to gather data about the process and outcomeof care in organizations in compliance with thestandards, the opportunities to conduct such re-search will increase.

Even as standards are being validated, the Fed-eral Government, State governments, and privateorganizations could take more interest in devel-oping and encouraging the use of consistent setsof standards for specific services and conditions.This step could increase consumers’ access toscope of services information, as well as to hos-pitals with at least a minimal level of resourcesfor conditions affecting them. Some consumersdo not have access to scope of services informa-tion, because available guidelines are not applied

204

Table 9-7.—Characteristics of External Standards and Guidelines for Hospitals:Overall Accreditation/Certification and Specific Services

Survey byVoluntary independent Publicly Ease of access

Validated or mandatory observers available to information

Overall hospitai accreditationlcertlfication:JCAHO . . . . . . . .Some studies;

generally not

HCFA . . . . . . . .f’Jo

Standards and guidelines for specificNeonata/ intensive care services:

AAPIACOG . . . . Level Ill/outcomestudies

States . . . . . . . . Results differ byState

Cancer care:ACS . . . . . . . . . . NO

ACCC . . . . . . . .No

Voluntary Yes Accreditation Accreditation his-history tory easy; other

informationdifficult

Mandatory for Yes Yes Difficultparticipatinghospitals

services:

Voluntary

Varies

No No Difficult

Ohio, some other Ohio, some others DifficultStates; not byAA PIACOG

Voluntary

Voluntary

Yes

To begin

Yesa

Yesb

Fact of approvalrelatively easy;more detailedinformationdifficult

Fact ofmembershiprelatively easy;actualadherence tostandardsdifficult

Emergency services:JCAHO . . . . . . . No Mandatory for Yes No, except through Difficult

participating some States andhospitals willing hospitals

ACEP/ENA ., . . No Voluntary No No NAAMA . . . . . . . . .t’Jo Voluntary No No NAStates . . . . . . . . No Both ? Some Varies

Trauma:ACS . . . . . . . . . . bJo Voluntary Under consideration No NAStates . . . . . . . . Some studies

but poormethodologically Both Some Probably Varies

Abbreviations: AAP/ACOG = American Academy of Pediatrics and American College of Obstetricians and Gynecologists; ACCC = Association of Community CancerCenters; ACEP/ENA = American Coiiege of Emergency Physicians and Emergency Nurses Association; ACS = American Coiiege of Surgeons; AMA= American Medical Association; HCFA = Heaith Care Financing Administration; JCAHO = Joint Commission on the Accreditation of Heaithcare Or.ganizations.

aList of approved hospitais.bList of member hospitals, Wtro may or may not foliow ACCC guidelines.

SOURCE: Office of Technology Assessment, 19S8.

205

by the organizations that developed them or byregulatory bodies. The development of standardshas been slowed by professional rivalries, as wellas by financial concerns (194,627), and lack of evi-dence about which requirements are valid. Lessthan half of all States have designated a Statetrauma center program, for example. Concertedefforts to develop consistent standards will beneeded to overcome these problems.

In conclusion, considerable research is neededto validate accreditations/certifications for hos-

pitals overall and external standards for specifichospital services. At present, accreditations, cer-tifications, and approvals by independent bod-ies of experts seem to be a necessary, but not suffi-cient, indicator of a minimum standard of qualityfor hospitals overall and for some specific serv-ices. At the same time that research to developmore valid standards is being conducted, Stateand Federal governments could encourage the useand dissemination of information about hospitals’compliance with existing standards.