Retrospective Study of Chest Pain Cases Presenting to a Chiropractic Teaching Clinic: A Preliminary Feasibility Study

Monica Smith, D.C., Ph.D.,  Palmer Center of Chiropractic Research, Davenport, Iowa, Matthew Ellerbrock, D.C.,  Columbus Grove, Ohio, Khaled Khorshid, M.B.B.Ch., M.S., D.C.,  Kentuckiana Children’s Center, Louisville, Kentucky, and Shellee Handley, D.C., Palmer College of Chiropractic, Davenport, Iowa

 

 

RECENT STUDIES SUGGEST that up to one half of chest pain presentations to various primary contact ambulatory medical settings may be musculoskeletal in origin. It is not clear at this time how and to what extent chest pain sufferers utilize chiropractic health care services. Patient charts at a chiropractic teaching clinic were retrospectively reviewed to determine the degree to which documentation regarding the presentation of chest pain was consistently recorded at nine data capture points within the patient’s clinical record. Results from this preliminary feasibility study indicate that of the 108 randomly drawn adult patient records, 24 potential chest pain “cases” were identified, only one of which (4%) was also missing a key component of the patient record. Of the 84 potential non-chest pain “controls” identified, 33% (n=28) were completely missing one or more important elements of the record. It is important to document the extent to which chest pain sufferers are seeking out chiropractic care as a first, or later, point of contact with the health care system, and the degree to which such care seeking is appropriately documented in the clinical records. To the extent that accurate and accessible clinical data will allow reliable tracking of the patient during the entire chiropractic health care encounter, the appropriateness with which such cases are managed in chiropractic teaching clinics or other settings may also be monitored. Multisite collaborative studies across chiropractic teaching clinics, and in practice-based settings, would greatly improve our understanding of the nature of this important professional issue. (JNMS: Journal of the Neuromusculoskeletal System 8:67-75, 2000)

Key words: Academic health centers, Chart review, Chest pain, Chiropractic, Documentation compliance, Feasibility study, Prevalence, Teaching clinics.   

 

According to data from the 1997 National Hospital Ambulatory Medical Care Survey, chest pain is the fourth most common presentation to emergency rooms, constituting 2.9%, or 2.8 million visits, out of the total 95 million visits to U.S. emergency departments annually (1). Results from a 1977 retrospective chart review reported a rate of 68 patients presenting with chest pain per 1,000 family practice patients (2). Extrapolating that rate to 1997 National Ambulatory Medical Care Survey estimates of 787.4 million total visits to physician offices in the United States (estimated three annual visits per patient) (3), chest pain presentations to a medical physician office setting any constitute approximately 17.9 million patients annually. Chest pain has social consequences in terms of disability, physician visits, cardiac medication usage, repeated and costly hospital admissions, and cardiac catheterizations (4-6).

 

REVIEW OF LITERARTURE: MUSCULOSKELETAL CHEST PAIN

Clinical health services studies in the 1970s and 1980s documented that a substational number of chest pain presentations were largely of indeterminate origin, based on the diagnostic procedures and criteria widely in use during that time period. One early study indicated that in 157 of 200 chest pain patients, the pain originated from nonorganic sources, most likely of the so-called “musculoskeletal type” (7). In a 1977 retrospective chart review of 109 family practice patients with complaint of chest pain, Blacklock (2) placed each chest pain patient into one of the three groups, according to the final etiology of the chest pain as recorded in the chart: Chest pain of organic etiology with signs (e.g., lobar pneumonia, myocardial infarction (MI) fractured ribs); chest pain of organic etiology without sign (e.g., acute bronchitis, hiatal hernia without radiographic confirmation, angina pectoris); and chest pain of unproven etiology (e.g., undiagnosed, intercostal muscle strain, chest wall pain, chest pain secondary to anxiety). Notably, 50% of all chest pain patients remained of “unproven etiology” even after 6 months of follow-up.

Levine and Mascette (8) prospectively evaluated 62 adult chest pain patients referred for coronary angiography to Walter Reed Army Medical Center Cardiology Service. Using a systematic physical exam protocol to identify musculoskeletal sources of chest pain, the investigators found that chest pain was reproduced on physical exam in 11% of these referred patients. The authors suggested that demonstration of musculoskeletal tenderness that reproduces chest pain, combined with noninvasive findings suggesting low probability of coronary artery disease, may be useful in decreasing incidence of unnecessary invasive cardiac evaluation, and appropriately directing initial therapy.

By the 1990s, the literature reflected an ever-growing awareness that, for chest pain presentations to primary contact medical settings (e.g., emergency and urgent care, physicians’ offices), a substational proportion may be musculoskeletal in origin. By one estimate, 14.7% of chest pain presentations to emergency rooms (ERs) may be musculoskeletal (9). A more recent observational study of chest pain patients presenting to the ER of an urban U.S. tertiary care hospital may pose far higher estimates (10). Current techniques for accurately evaluating and triaging chest pain presenters are inadequate, and risk-driven liberal admission policies contribute substantially to an estimated cost of $10 to $13 billion per year to rule out MI in low-risk patients, as noted by Tatum and Colleagues. Their investigative team developed a comprehensive, vertically integrated clinical strategy for evaluating chest pain patients as part of an ongoing, continuous quality improvement (CQI) process (10). This strategy consists of rapid clinical evaluation, and on the basis of ECG, character of chest pain, and history, assigns patients to one of five levels, as indicated in table 1. Tatum et al. report that of 1,187 consecutive patients, over one half were assigned to levels 1 through 4. Although not specifically reported as a statistic, a reasonable inference may be made that the remainder of this patient sample (over 40%) were assigned to level 5 (“noncardiac” chest pain) by the investigators, based on the methods reported in this study.

Up to 49% of ambulatory chest pain presentations to family practice clinics may be diagnosed as musculoskeletal, with an additional 10% remaining undiagnosed (11). A 1993 prospective study reported on 40 consecutive patients with chest pain presenting to an Internal Medicine Clinic for evaluation of noncardiac chest pain as a follow-up for patients having undergone coronary angiography (12). Each patient in this study underwent a complete rheumatologic exam, which determined that 30% of the patients had fibromyalgia and 10% had costochondritis.      
TABLE 1. Tatum el al.’s Acute Cardiac Evaluation and Triage Levels


Level

Characteristics

Level 1

High probability of acute ischemic injury

Level 2

High probability of acute ongoing UA and potential for ischemic injury

Level 3

High probability of UA but low probability of AMI

Level 4

Low probability of AMI and low/moderate but significant possibility of UA

Level 5

Noncardiac chest pain defined by a clear diagnosis of chest pain that is not related to myocardial ischemia

UA, unstable angina; AMI, acute myocardial infarction.

 

            The proportion of chest pain due to musculoskeletal causes would reasonably be expected to substantially vary by setting. More severe cases of chest pain are more likely to present to an ER for immediate evaluation either as a referral from a health care professional (based on signs/symptoms) or because patients may self refer there based on their own perceived severity of the symptoms. Based on the literature available, this author was unable to discern the current prevalence of chest pain presenters to a chiropractic health care setting (13, 14).

Pediatric chest pain, though not as prevalent as adult, is noteworthy because children are more likely to have their chest pain remain undiagnosed for a considerable length of time. Also, children with chest pain consume considerable health system resources through multiple ER and physician visits. Accounting for 0.2%-0.6% of pediatric patient visits to ERs or outpatient clinics, chest pain is reported as a frequent presenting symptom of teenagers, accounting for 650,000 of their primary care visits annually (15-23).

Idiopathic chest pain (no organic cause, no psychological factors found to explain symptoms) is perhaps the most common final diagnosis for chest pain in adolescents and children, accounting for 21%-39% of cases in various studies; and when a specific cause is determined, the most common cause is musculoskeletal, in 20%-30% of cases (15-23).

Perhaps one of the more cited and established experts in the area of pediatric/adolescent chest pain is Steven Selbst, from Children’s Hospital of Philadelphia. From 1979 to 1981, Selbst (24) conducted a retrospective study of adolescent chest pain presentations to a pediatric emergency room, to establish a chest pain occurrence rate per patient visit of 0.23%. Of those chest pain cases, 36% were of acute onset (1 day or less); 33% had experienced their chest pain for 2 days to 1 month; chest pain was present for more than 1 month in 8.6%, and for more than one year in 7.8%. Sixteen percent of adolescents with chest pain had been evaluated in the same ER more than once for chest pain; 7.5% had been in the ER for leg pain, back pain, or abdominal pain at some other time. Idiopathic was the most common chest pain diagnosis (28%), followed by functional/anxiety (17%), musculoskeletal (15%), and costochondritis (10%). Selbst (25) next conducted a prospective study from 1984 to 1985 using the same ER setting.  The data show a similar, yet slightly changed pattern: the most common causes of adolescent chest pain admissions to the ER were idiopathic (21%), musculoskeletal (15%), and costochondritis (9%). Chest pain caused 30% of these children to stay out of school and 31% to awaken from sleep. Selbst notes that children with chest pain who are prospectively evaluated in a systematic fashion tend to have a slightly higher incidence of organic illness.

Recent studies further discriminate between specific diagnostic entities embedded within the larger class of musculoskeletal chest pain disorders in children. For instance, slipping rib syndrome (abnormal mobility of the 8th, 9th, or 10th rib) has been identified as causing abdominal pain and is often confused with intra-abdominal disorders, but this syndrome can also cause chest pain (21, 26). As well, sources of adolescent chest pain heretofore unknown or not well understood are of growing and crucial interest. For example, increasing clinical attention is being directed toward establishing diagnostic protocols for rapidly identifying juvenile chest pain, coronary vasospasm, and even MI associated with illicit and licit drug use, which is an issue of high priority in the midst of epidemic substance abuse among contemporary youth (27, 28).

In a comprehensive and current review on the topic of adolescent chest pain, Kocis (29) summarizes the organ systems responsible for causing chest pain in children. Noted is the variability across study findings that reported rates of idiopathic chest pain ranging from 12% to 85% of chest pain cases, and musculoskeletal ranging from 15% to 31%. Again, given the likelihood that patients may either self-select or be referred to various settings depending on the severity of their symptoms (or the extent to which prior diagnostic workup has already been completed), a large portion of the above noted variability in rates of patients with chest pain may be principally due to differences in study setting.   

Based on current available evidence in the literature, it is unknown what proportion of adult or juvenile chiropractic patients in various settings (e.g., field practitioner’s offices, teaching clinics) present with chest pain, either as chief complaint or associated findings. The purpose of this study of to explore the feasibility of gathering useful data on adult chest pain presentations to a typical chiropractic teaching clinic, using retrospective review of existing patient records. Only adult patient records were included in this study, becausespecialize pediatric clinical records (as are used in this teaching clinic) did not allow for the employment of a standardized methodology for identifying potential chest pain cases across age cohorts.

 

METHODS  

The investigators (three faculty, one intern) reviewed the patient record-keeping system in place at the study site, a Midwestern chiropractic teaching clinic. A draft form for abstracting study data was developed by identifying the full scope of record elements (94 variables) likely to yield potentially useful information regarding the presentation of chest or related pain. Following a brief “pilot” testing of the form and subsequent modification, a final hard copy data abstraction instrument (see Appendix) was employed to abstract data from 115 clinical records that were randomly drawn from all new cases presenting at the teaching clinic from June 1, 1998 to December 1, 1998. Seven of the initial randomly drawn records used specialized pediatric clinical records, and were dropped from the analysis. This yielded 108 standardized adult patient records for inclusion in this feasibility study. All data abstraction was performed by the student clinician member of the project team, in partial fulfillments of requirements for receiving Research Honors.

Although the data abstraction protocol yielded 94 potentially useful variables, only those 10 that explicitly referenced “chest pain” are considered and reported here. Records were viewed to determine the degree to which documentation regarding the presentation of chest pain was consistently recorded at the 10 following data capture points, within six (a-f) relevant elements of the patient’s clinical record:   

a- The Patient Intake Form (self-administered by patient)

1- Patient’s Health Problem List: containing up to four problems listed by patient in order of severity.

2- Patient’s response to prompt “Check box if you are now experiencing or have had previously: "Chest Pain"

3- Patient’s response to prompt “Check box if you are now experiencing or have had previously: " Pain Over Heart"

b- Cardiac Screening Questionnaire Form (self-administered by patient)

4-Patient’s response to prompt: "Check box if: You ever experience chest tightness, heaviness, pressure, or pain"

c- Narrative of the Patient’s History (x\extern/clinician report)

5- Doctor’s Notes (e.g., following up in interview whether “chest tightness” as self-reported by patient was past or current, querying patient as to possible provocations for that pain, etc.)

d- Narrative of the Patient’s Physical Exam (extern/clinician report)

6- Doctor’s Notes: Respiratory System Regional Exam (e.g., notation of whether exam maneuvers were performed to better differentiate the “chest pain” or “chest tightness”, etc.)

7- Doctor’s Notes: Cardiovascular System Regional Exam (e.g., notation of whether exam maneuvers were performed to better differentiate the “chest pain” or “chest tightness”, etc.)

8- Narrative Summary of Entire Physical Exam (explicit notation of “chest pain” or “chest tightness”, and findings of differential exams, follow-up studies, etc.)

e- Pain Diagram (extern/clinician report)

9- Anterior and posterior body sketches (chest or shoulder shaded as general pain areas)

f- Patient Management Plan Form (extern/clinician report)

10- ICD-9-CM code 786.5: Chest Pain (noted as primary, secondary, or tertiary diagnosis)

All data for this study were managed and analyzed in hard copy form; no statistical packages were employed during the analysis.

 

RESULTS

            Results from this preliminary feasibility study indicate that data extraction via retrospective chart review yielded one complete observation (containing 94 variables) every 10 minutes on average (ranging from 4 to 15 minutes per chart). As represented in Table 2, compliance with the record-keeping protocol at this teaching clinic varied by each component of the patient record. While the compliance rate for patient record documentation exceeded 85% for most components of the records sampled for this study, compliance with completing the Pain Diagram was lower, with a compliance rate of approximately 75%. As well as, a post-hoc review of the final data abstraction protocol used in this preliminary feasibility study revealed another problem. The data abstraction instrument allowed abstraction to mark the form as “positive” if either the chest or shoulder area were marked on the Pain Diagram in the patient’s chart, rather than coding each of these two areas separately. For these reasons, although four cases were potentially “positive” for chest pain based on the Pain Diagram component was excluded from further analysis. However, it should be noted that two of these potential Pain Diagram “cases” were also “positive” for chest pain in other patient record elements.



Analysis of the variable set was limited to the nine remaining data element points contained in the Patient Intake Form (three items), Cardiac Screening Questionnaire (one item), History (one item), Physical Exam (three items), and Working Diagnosis (1 item). Of 108 records reviewed, 84 were entirely absent of chest pain indicators in any of these nine patient chart data points, while 24 potential chest pain “cases” were identified, based on a positive chest pain finding at any of the nine data points.

As indicated in Figures 1 and 2, the lagging” of potential chest pain “cases” was based almost exclusively on those elements of the record filled out by the patient at intake. Only one of the 24 “cases” had a notation of chest pain also appearing in the clinician’s accompanying narrative (in the patient history). Most frequently, the patient-reported chest pain “flag” appeared solely in response to the query prompt in the Cardiac Screening Form (in 13 of 24 cases). In 8 of the 24 cases, the patient answered affirmatively to prompts in both the Patient Intake Form and the Cardiac Screening Form. In the remaining three cases, the patient only responded positively to the Patient Intake Form prompts, but not to the Cardiac Screening Form prompt.

It should also be noted that 33% (n = 28) of the non-chest pain “control” patient chart were completely missing one or more important elements of the record (e.g., regional physical exam not performed or noted for that patient; patient history missing from record; no ICD-9-CM diagnosis noted in care plan, etc.). One of the 24 potential chest pain “cases” identified in this chart review, only one was also missing a key component of the patient record.

Interestingly, while “chest pain” was never noted as a presenting complaint on the patient-reported Health Problem List for this entire sample of 108 patient records, a complaint of“shoulder pain” or “between shoulder pain” was noted on the patient’s problem list in 5 of the 24 “case” records (21%), but in only 8 of the 84 “control” records (10%).

 

DISCUSSION/CONCLUSION

The intended purpose of this study bears repeating here. This was not intended to be a study of documentation compliance, and the “compliance statistics” reported herein do not meaningfully add to that particular body of evidence in the health care management sciences (i.e., as might be found in the clinical/risk management or quality improvement literature). If this were a “compliance study,” then it would have employed a methodology to allow a clearer determination of why components of the patient record were missing or inadequately documented, if in fact they were. For instance, a clinical protocol in a teaching clinic might allow the student intern to conduct the full diagnostic workup over multiple visits rather than requiring that all exams, tests, and procedures be completed and documented by the intern during the initial visit. Teaching clinicians who have aided the fundamental skills development of a novice intern can understand why such a protocol is perfectly reasonable for a teaching clinic, out of respect for the patient’s time. During such initial visits, the clinician mentor may expertly shortcut through key clinical procedures and a working diagnosis to ensure that patient care and safety are not compromised during the intern training process, and the patient record during this time frame may appear abbreviated (on “incomplete,” to a chart abstractor). Over the next several visits, the remaining components of the patient record would be filled in, as the clinician mentor facilitates and guides the intern’s self-directed clinical discovery process through a comprehensive differential diagnostic workup and documentation of same. This is how good doctors are shaped, and we would expect this perfectly appropriate teaching process to be reflected in the patient record in such a setting. If, for whatever reason, the patient does not return following his or her initial visit(s), then the patient’s record on file would likely show as “complete” those components conducted during the initial visit(s), which may more likely be self-administered patient query instruments. This is somewhat consistent with the general pattern of completed documentation noticed in the current study.

Although this is not a “documentation compliance study” per se, it is interesting to consider the preliminary “documentation compliance rates” reported here, in the context of findings reported in other studies. For instance, a recent study examined physician compliance with various documentation rules regarding history and physical examination, based on retrospective blinded chart review of nontramautic chest pain patient records at 12 metropolitan emergency departments (30). The American College of Emergency Physicians’ (ACEP) chest pain clinical policy rules for physician documentation were the standard against which subject’s compliance was compared. According to the ACEP clinical policy, the essential components of the documented history included character of pain, associated symptoms, and past medical history. For physical examination, the ACEP required components are vital signs (including temperature), a cardiac examination, and a pulmonary examination. In that study, compliance with documentation rules ranged from a low of 87% (associated symptoms component) to a high of 99% (pulmonary examination). Cumulative compliance scores (i.e., all components complete) were 78% for the complete history and 88% for thecomplete physical exam. The methodology employed in this feasibility study does not allow for an accurate determination of the true documentation compliance rate for this chiropractic teaching clinic sample, given the likely effect of the “teaching factor” on that estimate and other possible factors not considered. But it may be reasonable to posit, given the direction of bias that has likely been introduced by the teaching factor, that the “true” compliance rate for this study sample would be higher than that reported here.

Documentation compliance was used in this chest pain feasibility study in order to explore the potential effect of that compliance on estimated rates of patients with chest pain. Record-keeping compliance dictates the appropriate denominator to be used in such studies, as inappropriately counting “missing” data as “negative” data introduces a downward bias into estimated rates of chest pain. This issue is clearly illustrated by reconsidering the data presented in Table 2 and Figures 1 and 2. Using the 24 potential chest pain “cases” identified across the various record components as a baseline, and considering all 108 patient records reviewed in this study, one of three assumptions about the 40 incomplete records might be made: 1) that had those records been complete, they would all have contained positive chest pain indicators (“cases”), or 2) they would have been negative for chest pain (“control”), or 3) some mix in between. The two extreme polar assumptions would generate chest pain presenter rates of 60% (64/108) or 22% (24/108), which expectedly fall outside of, and end-anchor, the range of rates reported in Table 2. However, these are merely speculative assumptions-the only “facts” we have to go on are the fully completed records, or fully completed record components. Using only fully complaint records seriously curtails the sample size available in this already small study (to n = 68), making any such generated estimates inherently unstable. It would have been possible in this small feasibility study, and necessary in a larger retrospective case-control study, to compare compliant records to noncompliant records, in order to determine whether the complaint records used for analysis were representative of the larger sample, and if not, the direction and nature of the resultant bias. However, the information to be gained from a study of such small size and limited scope as this, did not justify the added analytical work. It is important to reiterate that this was conducted as a feasibility study intended to help guide the development of future research in this area, and that any data reported here (e.g., rates of documentation compliance or chest pain “prevalence”) should be interpreted with cautious speculation.

The potential chest pain indicators explored in this study raise questions about the appropriateness of these various indicators for defining a “case”. As revealed in this study, the most reliable indicators, in terms of documentation compliance and yield only, were the patient’s self-reported responses to prompted queries contained in the patient intake form and cardiac screening questionnaire. The use of prompted queries, or of patient self-report versus clinician-report measures, poses conceptual and methodological issues deserving careful consideration in and of themselves. As well, the patient intake and cardiac screen had queries used in this teaching clinic do not distinguish whether the chest pain is current or past (i.e., do you now or previously…; do you ever experience…). Whether these queries, as currently written, might be useful in defining a chest pain “case” would clearly depend on the purpose of the study. While these queries may serve to measure overall gross prevalence of chest pain among chiropractic patients, the wording of the queries does not allow for the identification of new “incident” cases. Depending on the research question of interest, it may also be important to determine whether a patient with current chest pain considered it to be one of their principal reasons for seeking chiropractic care, a lesser reason, or even whether it was a revealed comorbid condition but not necessarily an overt reason for the patient’s seeking care from a particular type of provider.

The above discussion outlines some of the challenges inherent in employing retrospective chart review, such as was demonstrated in this study. This is not to say that such a methodological approach is fraught with insurmountable obstacles, but rather that careful attention needs to be paid to the design of such studies to ensure the gathering of useful data (31). The increasing trend towards standardized electronic record keeping in the clinical setting raises promise for improving documentation compliance (32) and for further enhancingthe efficiency and potential utility of retrospective review for clinical health services studies.

The preliminary exploration does suggest that when chest pain appears as presenting or associated compliant reported by the patient during intake, this finding may be typically absent from that patient’s history and physical exam clinician notations, and may never appear in the list of differential or comorbid diagnoses for these same patients. As ICD-9-CM diagnostic codes are those most likely to be submitted for third-party payor reimbursement purposes, then health services research using claims-based data extraction approaches may likely suffer from consistent under-reporting of important patient condition codes such as chest pain, when compared to patient’s self-reported presentation/intake information. As ICD-9-CM codes are typically used to characterize the chiropractic health care encounter in claims-based studies of health care delivery, this study is an important first step toward documenting the extent to which such claims-based studies may be severely biased by not accurately representing the full complexity and severity of health conditions represented in the chiropractic patient population. Again, depending on the research question and the issues of chief concern, a prospective approach to data collection may serve more useful than the retrospective approach explored in this feasibility study.

It is important to document the extent to which chest pain sufferers are seeking chiropractic care as a first, or later, point of contact with the health care system, and the degree to which such care seeking is appropriately documented in the clinical records. To the extent that accurate and accessible clinical data will allow reliable tracking of the patient during the entire chiropractic health care encounter, then the appropriateness with which such cases are managed in chiropractic teaching clinics or other settings may also be monitored. Given the dearth of information available, further work on this topic is needed. In particular, multisite collaborative studies across chiropractic teaching clinics, and practice-based settings, would greatly improve our understanding of the nature of this important professional issue.


 


REFERENCES

     

    • Nourjah P. National hospital ambulatory medical care survey: 1997 emergency department summary. Advance data from vital and health statistics. No. 304, Hyattsville, MD: National Center for Health Statistics, 1999.
    • Blacklock SM. The symptom of chest pain in family practice. J Fam Pract 1977; 429-433.
    • Woodwell DA. National ambulatory medical care survey; 1997 summary. Advance data from vital and health statistics, No. 305, Hyattsville, MD: National Center for Health Statistics, 1999.
    • Potts SG, Bass CM. Psychsocial outcome and use of medical resources in patients with chest pain and normal or near-normal coronary arteries: a long-term follow-up study. Q J Med 1993; 86(9):583-593.
    • Selker HP. Coronary care unit triage decision aids: How do we know they work? Am J Med 1989; 87:491-492.
    • Weingarten SR, Ermann B, Reidinger MS, et al. Selecting the best triage rule for patients hospitalized with chest pain. Am J Med 1989; 87:494-500.
    • Cope RL. The psychogenic factor in chest pain. Tex Med 1969; 65(5):78-81.
    • Levine P, Mascette A. Musculoskeletal chest pain in patients with “angina”: a prospective study. South Med J 1989; 82(5):580-585.
    • Jonsdottir O, Gottskalksson G, Porgeirsson G, Theodorsson A. Causes of typical and atypical chest pain. Iceland Med J 1990; 76:457-481.
    • Tatum JL, Jesse RL, Kontos MC, Nicholson CS, Schmidt KL, Roberts CS, Ornato JP. Comprehensive strategy for the evaluation and triage of the chest pain patient. Ann Emerg Med 1997; 29:116-125.

 

 

 

Orion Diagnostic & Chiropractic Center - Dr. Khaled Khorshid
Chicago Ridge Medical Center
9830 Ridgeland Ave. Suite 5 - Chicago Ridge, IL 60415
Tel: (708) 288-2239

home / about us/ services / research / consultation/ testimonials / contact

Powered by Shurouq
© 2009 althealthcarecenter.com
All rights reserved
.