Evaluation of an EMS regional referral system using a tracer methodology

ORIGINAL CONTRIBUTION

Evaluation of an EMS Regional Referral System Using a Tracer Methodology Joanne Egges, BA Bernard J. Turnock, MD, MPH. Springfield, Illinois

To,improve patient care, cost effectiveness, and resource utilization, the Illinois EMS Program attempted to regionalize emergency medical care servies by identifying referral centers for critically ill and injured patients. The performance of one referral region was evaluated applying a tracer method designed to track patients whose clinical conditions, as determined by a clinical panel, required treatment at a designated regional center. The proportion of patients reaching the appropriate centers suggests that the regional referral system under study appropriately moved only about one half the patients with the tracer conditions to the designated referral center. Additional patients were transferred to hospitals with greater emergency department capabilities, although these hospitals were not formally designated centers. Except for low birth weight, factors that might be associated with referral patterns indicate no significant differences between those patients at the regional center and those treated elsewhere. Egges J, Turnock BJ: Evaluation of an EMS regional referral system using a tracer methodology. Ann Emerg Med 9:518-523, October 1980. emergency medical services systems, evaluation; referral system, EMS, evaluation I NTRO DUCTI ON

We attempted to evaluate aspects of regional referral networks in the Illinois Emergency Medical Services (EMS) Program. An early goal of the Illinois EMS Program was identification of hospitals as regional referral centers for critically ill and injured patients. 1 Boyd2 has argued that a regionalized approach, with a continual upgrading of trauma and emergency medical capability where substandard resources exist would improve patient care and produce additional benefits, such as cost effectiveness and improved resource utilization, by eliminating unnecessary duplication of efforts, monies, and medical manpower. Two studies on vehicular trauma by Boyd et al2, 3 and one by Mullner and Goldberg 4 have suggested that the Illinois Trauma System was successful in redistributing patients and in improving emergency medical care. Criticisms of Boyd's studies 5-7 have created confusion and skepticism as to the actual success of the system. Mullner's study 4 demonstrated decreased case fatality at trauma centers. Although this study also showed an overall increase in the number of patients treated at the trauma centers when patients were grouped by severity, the increase was significant only for those with minor injuries and not for the critically injured patients for whom the system was designed, s It remains to be proven that the Illinois EMS system, which was based on and incorporates the trauma program, has indeed directed critically ill and injured patients to the designated regional centers and in so doing has improved their outcome, s From the Illinois Department of Public Health, Division of Emergency Medical Services, Springfield, Illinois. Address for reprints: Joanne Egges, 1821 22nd Street, Apartment 206, Boulder, Colorado 80302.

9:10 (October) 1980

Ann Emerg Med

518/33

Our objective was to e v a l u a t e the EMS regional referral system in one r e f e r r a l r e g i o n to d e t e r m i n e whether patients were reaching the appropriate regional center. The Illinois Department of Public Health, Division of Emergency Medical Services and Highway Safety, as the lead agency responsible for the statewide program, plans to use the results to identify problems with regional ref e r r a l n e t w o r k s , to aid in f u r t h e r development or modification of the statewide system, and to provide assistance to the hospitals involved in planning and delivering emergency medical care.

METHODS The study followed five t r a c e r conditions - - multiple trauma, head t r a u m a , s p i n a l cord i n j u r y , b u r n s a n d low b i r t h w e i g h t i n f a n t s , t h r o u g h the s y s t e m to d e t e r m i n e whether patients were treated in the regional centers designated for their condition. The study design uses the t r a c e r method a d v o c a t e d by Kessn e r 1° as a model, although the lack of m a t u r i t y in EMS systems and systems research made several modifications necessary. There existed no specific guidelines for when a p a t i e n t should be transferred. A clinical panel of four physicians was given the responsibility of defining each tracer condition for the purposes of this study. They b a s e d t h e i r d e f i n i t i o n s on t h r e e p r i m e c o n s i d e r a t i o n s : 1) t h e definitions should include only those patients who require t r e a t m e n t at a reg i o n a l center; 2) t h e t r a c e r cases with a high or low probability of survival regardless of the t r e a t m e n t received should be excluded, thus focusing on those p a t i e n t s for whose benefit the system was designed; and 3) the definitions should be usable by nonphysician data collectors to identify tracer cases. The panel reviewed the current literature concerning the identificat i o n and t r e a t m e n t of p a t i e n t s in each t r a c e r category. In a d d i t i o n , protocols used by o t h e r s t a t e s to select the a p p r o p r i a t e t r a u m a patients and high risk infants for transfer were a s s e m b l e d and analyzed. Each physician was given the opportunity to respond to the criteria and to suggest additional criteria. Following is a general summary of the tracer definitions prepared by this panel. M u l t i p l e T r a u m a - Patients over 12 years of age with an injury severity score of 20 to 40. ~,~2

34/519

Table 1 SAMPLING RESULTS, ILLINOIS TRACER STUDY No.

Case Disposition Possible tracer condition - identified from ED log and PAS Reviewed hospital chart Abstracted hospital chart Did not meet study criteria Met study criteria

Head TraumaPatients over 12 years of age sustaining head injuries resulting in loss of consciousness for one hour or more after arrival at the hospital and/or a neurological deficit as a result of an injury directly to the b r a i n . P a t i e n t s w i t h a chronic or subacute subdural hematoma, a hygroma, a lacerated brain, a brain stem contusion, or a drug/alcohol intoxication are excluded. Spinal Cord Injury - Patients over 12 years of age who have sustained a fracture or dislocation of the spine, i n c l u d i n g the a t l a s or axis, w i t h e v i d e n c e of a n e u r o l o g i c a l deficit. B u r n - Patients over 12 years of age w i t h second or t h i r d d e g r e e b u r n s of 20% to 50% body surface area. Low-Birth-Weight I n f a n t - Infants with a birth weight between 1,000 and 2,000 gm (approximately 2 p o u n d s , 3 ounces to 4 p o u n d s , 7 ounces). Data collectors who had previous health-related education and experience were t r a i n e d as field staff to identify potential cases and abstract d a t a from the medical records. Potential cases were identified through emergency department logs, Profess i o n a l A c t i v i t y S t u d y (PAS) discharge summaries, and b i r t h logs. The medical records for each potent i a l tracer case were reviewed. For cases accepted by the field staff, the following information was abstracted onto data collection forms from the m e d i c a l record: p a t i e n t age, discharge diagnosis, method of transportation, number of transfers with time and location, and length of stay. The final step for acceptance as a tracer case was a review of the data collection forms by the e v a l u a t i o n staff and the clinical panel. The t a r g e t population for this study consisted of all patients meeting the tracer definitions who were

Ann Emerg Med

3,491 3,412 416 197 219

treated at and discharged from Region 3A hospitals. In general, this w e s t - c e n t r a l r e g i o n of I l l i n o i s is characterized by an urban-rural division: TM 55% of the population live in urban areas; the remaining 45% live in r u r a l a r e a s . The s t u d y design called for a complete sampling of the 22 Region 3A hospitals. The study period was 1975: six months for the four types of t r a u m a cases and one y e a r for infants.

RESULTS A total of 3,491 p a t i e n t s were identified. Approximately 2% of the records could not be located for review. Some 12% (416) of the cases were preliminarily accepted. Further consideration eliminated 197 of these cases; the majority were excluded as not meeting the severity criteria. Of the 197 cases excluded, only seven were not included because the medical record lacked complete study information. A total of 219 cases thus fit the t r a c e r condition definitions and were available for study (Table 1). Of these, 89 were t r a u m a cases and 130 were low-birth-weight infants. The data for low-birth-weight infants and for t r a u m a patients were a n a l y z e d s e p a r a t e l y due to differences in the s a m p l i n g p e r i o d and sample si~ze. The p r o p o r t i o n of t r a u m a patients in each tracer category treated at regional referral centers and at other hospitals is examined (Table 2). Fifty-two percent of the tracer cases eventually reached the approp r i a t e referral hospital; the remainder stayed at outlying hospitals. Transfer patterns of the 89 trauma p a t i e n t s were s u m m a r i z e d (Figure 1). When direct admissions to the ref e r r a l hospitals are excluded, only 41% of the patients went to the re g i o n a l centers. W h e n transferred, 90% (27 patients) were transferred

9:10 (October) 1980

Table 2 FREQUENCY AND PERCENTAGE OF TRACER CONDITIONS BY HOSPITAL TYPE Tracer Condition Head and multiple trauma Head trauma Multiple'trauma Burn Spinal cord injury Total

Regional f

Hospitals %

Other f

Hospitals %

Total f

36 (24)

51.4 (58.5)

34 (17)

48.6 (41.5)

70 (41)

(12)

(41.4)

(17)

(58.6)

(29)

10

83.3

2

16.7

12

0

0

7

100.0

7

46

51.7

43

48.3

89

Table 3 EMERGENCY CLASSIFICATION OF INITIAL TREATMENT HOSPITALS FOR PATIENTS TRANSFERRED TO THE REGIONAL CENTER Emergency Classification

Percentage (n = 30)

Comprehensive Basic Standby Total

0 46.7 53.3 100.0

Table 4 PERCENTAGE OF TRACER CONDITIONS BY EMERGENCY CLASSIFICATION OF HOSPITAL

Emergency Classification Comprehensive Basic Standby Total

Head and Multiple Trauma (n = 34) 26.5 50.0 23.5 100.0

Burn (n = 2)

Spinal Cord I njury (n = 7 )

Total (n = 4 3 )

0 0 100.0 100.0

71.4 1.4.3 14.3 1 00.0

32.5 41.9 25.6 100.0

Table 5 PERCENTAGE OF LOW-BIRTH-WEIGHT INFANTS BY WEIGHT CATEGORY AND HOSPITAL TYPE*

Hospital Type Regional center Other hospitals Total

Weight 1,000-1,599 1,600-2,000 gm gm (n = 51) (n = 79) 74.5 51.9 25.5 48.1 100.0 100.0

*Chi square significant at 0.05 level.

9:10 (October) 1980

Ann Emerg Med

within the appropriate time limits identified by the clinical panel. In addition to designations as regional centers, the general emerg e n c y d e p a r t m e n t capabilities of hospitals are categorized under regulations implementing Illinois state law (Public Act 76-1858) as comprehensive, basic, or standby. Comprehensive facilities are the most sophisticated; standby are the least. All of the cases transferred to the regional center came from standby and basic hospitals (Table 3); none was referred to the regional centers from comprehensive hospitals (Figure 2). The emergency classification of the treatment hospitals for patients who were not transferred to the regional center is shown (Table 4). O n e - f o u r t h of these p a t i e n t s remained at the lowest level of care, or s t a n d b y hospitals. Approximately one-fourth (23%) of these patients treated outside the regional center were transferred to a second outlying hospital (Figure 1). With one exception, they were transferred to hospitals with higher ranking emergency department capabilities. Factors that may be associated with those referral patterns were examined. None of these factors (injury severity scores for multiple and head trauma, length of hospital stay, patient age, and mode of prehospital transportation) showed any significant differences between patients at the regional center and those treated elsewhere. In the low birth weight category, there were 130 live born infants between 1,000 and 2,000 gm; 6!% were treated at the regional center for high risk infants; 39%, in outlying hospitals (Figure 2). When the infants who were born at the regional center are excluded, 49% of the infants were transferred to the center. Of these, 92% of the transfers were initiated within the established sixhour time limit. Infants not transferr.ed remained at the hospital in which they were born. When grouped by weight categories (Table 5), there are significant differences between patients treated at the regional center and infants treated elsewhere. There was also a s i g n i f i c a n t r e l a t i o n s h i p between birth weight and transfer (Table 6). Of the infants born at outlying hospitals, those in the lower weight category had a better chance of reaching the regional center than those in the higher weight category. The conditions reported for infants in the sample population are 520/35

listed (Table 7). These findings demonstrate a higher incidence of morbidity among infants at the regional center, and are consistent with the finding that the infants at greater risk, as measured by weight, are being transferred to the center. The a v e r a g e hospital stay is longer at regional centers than at other hospitals (Table 8).

Total Patient Sample 89 Patients who reached appropriate referral hospital

Patients who did not reach appropriate referral hospital

46

43

DISCUSSION

The Illinois EMS Program has a t t e m p t e d to improve e m e r g e n c y care for the critically ill and injured through regional referral systems. Results of this study suggest that such referral systems are not functioning in Region 3A as planned: only 41% of the trauma patients, or 52% including direct admissions, were treated at the appropriate r e gional center; 49% of the neonates were treated at the regional center, or 61% including births at the center. It should be stressed that these findings are based on a patient sample of five tracer conditions from one referral region and the findings are currently limited to these conditions in this geographic area. Our results may be typical of referral p a t t e r n s t h r o u g h o u t the United States. Unfortunately, there are no s t a n d a r d s for comparison, only isolated studies. Griffith's 14 study showed a higher percentage of patients treated at the appropriate level of care, but demonstrated a similar percentage for those needing transfer who were actually transferred: 50% compared with 41% in this study for trauma patients and 49% for low-birth-weight infants. Perhaps Gibson's s analysis is correct. "Current strategies in EMS in the public and private sectors disguise rather than change its structure." To confirm this theory, future studies would have to use a before-and-after approach. Results of this study have made Illinois EMS P r o g r a m m a n a g e r s aware that weaknesses exist in the regional referral system, at least in Region 3A. The question as to why patients are not getting to the "appropriate" regional referral centers r e m a i n s u n a n s w e r e d . Due to the enormous complexity of the interactions involved in the EMS system, t h e r e are m a n y possible explanations. First among the possible explanations would be flaws in the study design. The definitions for the five tracer conditions might have contrib-

36/521

Transfer admission

Direct admission

30

16

~

Transferred to another hospital

Remained at first hospital

10

33

~ ~ r a n s

Transferred within appropriate time period

ferred after appropriate time period

27

3

Fig. 1. Transfer summary for patient sample. Total Sample 130

Infants who reached high risk infant center

Infants who did not reach center

79

51

T r a n S e c t admission admission 49

30

initiated within appropriate time period

Transfer initiated after appropriate time period

45

4

Transfer

Transferred to another hospital 0

Remained at first hospital 51

Fig. 2. Transfer summary for low-birth-weight infants. uted to a nonrepresentative or inappropriate sample of those cases requiring care at the regional center. For example, the limitations of using only birth weight to identify infants with a high risk of morbidit~ are widely recognized. 15 Birth weight, g e s t a t i o n a l age, and p a t t e r n of i n t r a u t e r i n e g r o w t h give a much clearer picture of the infant's morbidity risk. If clearly defined treatment and transfer protocols had already existed, there would have been more objective criteria upon which the operation of the s y s t e m could be judged. Ann Emerg Med

A second weakness of this methodology is its dependence on a retrospective analysis of medical records which are notorious for their inconsistencies, inaccuracies, and missing information. The extent of this bias is unknown, but also unavoidable in a retrospective study design such as this. However, only seven cases of the 3,412 records reviewed were not included because the medical records lacked complete clinical information. A third weakness of the methodology is that it, like the EMS Reg i o n a l Plan, did not incorporate those who were transferred into the

9"10 (October) 1980

Table 6 DISPOSITION OF INFANTS BORN AT HOSPITALS OTHER THAN THE REGIONAL CENTER BY WEIGHT CATEGORY* Percentage

Disposition

1,000-1,599 gm (n =40)

1,600-2,000 gm (n =60)

67.5

36.7

Transferred to regional hospital Remained at other hospital Total

32.5

63.3

100.0

100.0

Total (n =100) °49.0 51.0 100.0

*Chi square Significant at 0.05 level.

Table 7 TYPES OF MORBIDITY AMONG LOW-BIRTH-WEIGHT INFANT BY TYPE OF HOSPITAL

Type of Morbidity Respiratory disease Other blood dyscrasia Major congenital anomalies Hemorrhage Infection Central nervous system Isosensitization Birth injury Othert

%

%

Regional Center (n =79)

Other Hospitals (n =51)

Total Percentage* (n =130)

74.7 40.0 21.5 16.5 2.5 2.5

27.5 21.6 13.7 2.0 3.9 0

56.2 31.5 18.5 10.8 3.1 1.5

1.3

2.0

1.5

0 30.4

2.0 5.9

0.8 20,8

* The total is greater than 100% because more than one illness could be reported for each infant. t "Other" includes metabolic disorders and major organ failures not included above.

Table 8 MEAN AND MEDIAN LENGTH OF STAY FOR INFANTS BY HOSPITAL TYPE*

Hospital Type

Length of Stay in Days Mean Median

Regional center

41

38

Other hospitals

20

19

*Excludes infants who died during their hospital stay.

9:10 (October) 1980

Ann Emerg Med

region from outside, an a d d i t i o n a l 18% (16 cases); or those who were t r a n s f e r r e d outside the region, another 20% (18 cases) who were potential but not confirmed cases. Another reason for the apparent lack of compliance with the intended referral network may be inadequate public education. Regionalization requires changes in patient attitudes as well as in the system. Findings from the I l l i n o i s EMS H o u s e h o l d Interview Survey TM indicated gaps in the public's knowledge of the availability of emergency medical services. M o r e o v e r , it i n d i c a t e d t h a t a strong belief persists that all emergency services should be available in all communities. Public understandi n g and acceptance of regionalized services had been slow in coming. Part of the problem could be the a t t i t u d e s and practices of v a r i o u s professional and i n s t i t u t i o n a l providers of emergency care in Illinois. Among a host of possible and/or pract i c a l explanations are issues such as the validity of the categorization efforts and general upgrading of services in all hospitals. Early efforts in c a t e g o r i z a t i o n h a d a n u m b e r of shortcomings. Often a regional center was designated on the basis of nonmedical factors such as location, political considerations, religious affiliations or general cooperative posture, while another equally qualified institution nearby remained undesignated. According to Gibson, 17 ' L . . w h e n c a t e g o r i z a t i o n becomes subject to political negotiation . . . the strategy is most vulnerable and has h a d least success." The Illinois ref e r r a l s y s t e m m i g h t have e x p e r i enced more compliance if a t r u l y community-based approach, as recommended by Tell, TM had been used. Hospital administrators might be resistant to categorization and referral networks for fear t h e y will decrease h o s p i t a l a d m i s s i o n s t h r o u g h the e m e r g e n c y d e p a r t m e n t which account for approximately one-third of all inpatient "days, resulting in lower occupancy rates which carry financial sanctions in many states, z7 Illinois hospitals also are subject to a categorization scheme (compreh e n s i v e , basic and s t a n d b y ) mandated by the Hospital Licensing Act. This is actually a self-designation effort t h a t often is not verified at the time of periodic hospital survey act i v i t i e s . While it is a p p a r e n t t h a t some hospitals in Region 3A have upgraded in recent years, it has been predicted for some time t h a t EMS regionalization funds would result in

522/37

a g e n e r a l u p g r a d i n g of hospital facilities and consequently could have a localizing effect such t h a t p a t i e n t s m a y not then need to be transferred. 5 A n o t h e r possible e x p l a n a t i o n for lack of appropriate referral p a t t e r n s is the existence of informal referral networks which serve to c i r c u m v e n t a n d complicate p a t i e n t flow patterns. Informal referral patterns include the use of n o n d e s i g n a t e d facilities for specialized care services, the referral of patients on the basis of t r a i n i n g or personal relationships with particular p h y s i c i a n s or i n s t i t u t i o n s , a n d even i n a d e q u a t e provider knowledge of the special care c a p a b i l i t i e s located w i t h i n either their own institut i o n s or o t h e r f a c i l i t i e s to w h i c h t h e i r p a t i e n t s m i g h t be r e f e r r e d . Some physicians are r e l u c t a n t to rel e a s e c o n t r o l of t h e i r p a t i e n t s to other institutions. Similarly, many i n s t i t u t i o n s place i m p l i c i t p r e s s u r e on p h y s i c i a n s to help e n s u r e h i g h censuses. I n n e a r l y all cases these difficulties are remediable by appropriate ongoing education directed toward the i n s t i t u t i o n a l a n d professional providers involved i n m a k i n g these inappropriate referrals.

CONCLUSION If the best t h a t can be said after the i n i t i a l tracer study is t h a t the referral system does not appear to be f u n c t i o n i n g as p l a n n e d , t h e n it still m u s t be d e f i n i t i v e l y d e m o n s t r a t e d t h a t the p l a n has a sound foundation i n i m p r o v i n g the q u a l i t y a n d outcomes of emergency care services. It m a y have been heroic, b u t unrealistic and somewhat simplistic, for

38/523

i n s t i t u t i o n s to a s s u m e t h a t categorization would lead with relative ease to e l i m i n a t i o n of inappropriate utiliz a t i o n a n d d u p l i c a t i o n of efforts, monies a n d medical manpower, a n d r e s u l t i n improved emergency care with demonstrable reduction in m o r t a l i t y , morbidity, a n d disability statistics.

REFERENCES 1. Boyd DR, Mains KD, Flashner BA: A systems approach to statewide emergency medical care. J Trauma 13:276-284, 1973. 2. Boyd DR, Mains KD, Flashner BA: A symposium on the Illinois trauma program: a systems approach to the care of the critically injured. J Trauma 13:275284, 1973. 3. Boyd DR, Pizzano WA, Romano TL, et al: Regionalization of trauma patient care: The Illinois experience, in Nyhus LM (ed): Surgery Annual. New York, Appleton Century Crofts, 1975, pp 41-52. 4. Mullner R, Goldberg J: The Illinois trauma system: changes in patient survival patterns following vehicular injuries. J A C E P 6:393-396, 1977. 5. Gibson G: Emergency medical services: regionalizing intents and localizing efforts, in Ginzberg E (ed): Regionalization and Health Policy. Department of Health, Education and Welfare, Public Health Service, Health Resources Administration, DHEW Publication No. (HRA) 77623, 1977, pp 85-98. 6. Willemain TR: The status of performance measures for emergency medical services. J A C E P 4:143-151, 1975. 7. Smock SN, Thall ML: Negative impacts of the model EMS system: it is time for an honest reappraisal. J A C E P 6:206208, 1977.

Ann Emerg Med

8. Mullner R: An assessment of the Illinois trauma program: an evaluation in

regional health planning. Unpublished thesis, Department of Geography, University of Illinois, Urbana, Illinois, 1976. 9. Final Report - A Plan for the Development of a Total Emergency Services System for the State of Illinois. DHEW Demonstration Contract HSM 110-72-345.

10. Kessner DM, Kalk CE, Singer J: Assessing health quality: the case for tracers. N Engl J Med 288:189-194, 1973. 11. Baker SP, O'Neill B, Haddon W, et al: The injury severity score: a method for describing patients with multiple injuries and evaluating emergency care. J Trauma 14:187-196, 1974. 12. Semmlow JL, Cone R: Utility of the injury severity score: a confirmation. Health Serv Res 11:45-52, 1976. 13. Current Population Reports: Population Estimates and Projection. Department of Commerce, Bureau of Census, Series P-25, No. 661, 1977. 14. Griffith JE: Facilities Categorization, paper presented at DHEW's EMS System Evaluation Symposium, Seattle, 1978. 15. Lubenceco LO: The high risk infant, in Schaffer AJ, Markowitz M (eds): Major Problems in Clinical Pediatrics, Philadelphia, WB Saunders, 1976, pp 1-8. 16. Peisert MA: Illinois Emergency Medical Services: Household Interview Survey. Southern Illinois University School of Medicine, Department of Health Care Planning. DHEW Demonstration Contract HSA 110-72-345, 1976. 17. Gibson G: How far have we come with categorization. Hospitals 51:97-102, 1977. 18. Tell R: Categorization: a community based approach. J A C E P 4:152-155, 1975.

9:10 (October) 1980