1

CLINICAL STUDIES

Alcohol Abuse: Comparison of Two Methods for Assessing Its Prevalence and Associated Morbidity in Hospitalized Patients ANNIEUMBRICHT-SCHNEITER,M.D., PATRICIASANTORA,Ph.D., RICHARDD. MOORE,M.D., M.H.s., Baltimore, Mary/and

PURPOSE To evaluate two methods for assessing the prevalence of alcohol abuse iu hospitalized patients based upon scores on standardized alcoholism screening instruments compared with diagnostic discharge data, and to determine the risk for comorbid conditions in patients who abuse alcohol. PA!rIIWTS AND MRTRoDS: Of 2,534 consecutive patients admitted to five adult inpatient services of an academic center, 1,964 were screened for alcohol abuse using the CAGE and the SMAST. Their discharge diagnoses were obtained and analymd for the presence of alcohol-related diagnoses and other comorbid conditions. RESULTS: A total of 1.4% of patients had a principal alcohol-related diagnosis (ARD), 6% had a secondary but no principal ARD, and 15% screened positive for alcohol abuse but had no ARD. The overall prevalence of alcohol abuse was 22.4%. Patients with a principal ARD had a bigher risk for dementia, chronic obstructive pulmonary disease (COPD), pancreatitis, sequelae of liver disease, and illegal drug abuse. Patients with a secondary ARD were at risk for 19 comorbid conditions, including pancreatitis, injury, pneumonia, COPD, and poly-drug abuse. Patients who screened positive for alcohol abuse but had no ARD were significantly more likely to have a diagnosis of hypertension, arrhythmia, breast cancer, or pelvic inflammatory disease. CONCLUSION: Discharge diagnoses done markedly underestimate the prevalence of alcohol abuse in hospitalized patients. Patients from the three groups are at higher risk for comorbid conditions, and secondary prevention of alcohol abuse can be achieved by routinely screening ev-

From the Department of Medicine (AU, RDM) and the Blades Center for Clinical Practice and Research in Alcoholism (PEG). The Johns Hopkins University, School of Medicine, Baltimore, Maryland. This work was supported by Contract ADA 281-86-0003 from the National Institute on Alcohol Abuse and Alcoholism. Dr. Moore is a Burroughs Wellcome Scholar in Pharmacoepidemiology. Requests for reprints should be addressed to Richard D. Moore, M.D., M.H.S., Johns Hopkins University School of Medicine, 1830 East Monument Street, 8th Floor, Baltimore, Maryland 21205. Manuscript submitted December 13. 1990, and accepted in revised form March 18, 1991.

110

August 1991

The American Journal of Medicine

Volume 91

ery patient using recognized ing instruments.

T

alcoholism

screen-

he association between alcohol abuse and increased morbidity and mortality in alcoholic populations has been well documented [l-lo]. Little is known, however, about alcohol-associated morbidity in patients in short-stay hospitals since physicians do not routinely screen their patients for alcohol abuse [ll]. Indirect measures of alcohol abuse, such as rates of liver cirrhosis or the per capita consumption of alcohol, have been cited to assess its prevalence. To more accurately assess the prevalence of alcohol-associated morbidity in patients admitted to short-stay hospitals, researchers at the National Institute on Alcohol Abuse and Alcoholism (NIAAA) developed a system using discharge diagnostic codes converted from the ICD-9CM criteria [12,13]. Using these criteria, a national survey of hospital discharges indicated that 4.9% of all discharged patients had an alcohol-related diagnosis (1.7% as the principal diagnosis and 3.2% as a secondary discharge diagnosis; [14]). Understanding alcohol-associated morbidity is essential to patient management and to developing effective treatment plans for these patients seen in the general hospital. It is essential to maintain a close consultative liaison between alcoholic treatment services and psychiatric, medical, and surgical staff in developing appropriate management of the medically ill, alcohol-abusing patient. Serious problems of diagnostic assessment, treatment planning, and patient compliance are inevitable without an adequate understanding of how alcohol abuse is associated with other morbidities. The use of standardized alcoholism screening instruments has repeatedly demonstrated that the estimated prevalence of alcohol abuse among hospitalized patients ranges from 15% to 30% [X-21]. This suggests that alcoholism screening instruments provide a substantially different prevalence of alcohol abuse in hospitalized patients when compared with discharge diagnoses. The difference between these two methods in estimating the prevalence of alcohol abuse and its associated morbidity in the same hospitalized population has not been measured previously. Several studies have docu-

ALCOHOL-ASSOCIATED

mented the prevalence of alcohol-associated morbidity in patients admitted to general hospitals [22-261; these studies, however, contained several methodologic weaknesses such as using small or nonrepresentative patient samples, restrictive diagnostic criteria, or nonstandardized instruments to assess the patients’ drinking histories. Therefore, this study of hospitalized patients admitted to a general, short-stay hospital has two specific aims: (1) to evaluate two methods for assessing the prevalence of alcohol-associated morbidity based upon the use of standardized alcoholism screening instruments compared with diagnostic discharge data; and (2) to quantify the risk for other comorbid conditions found in patients identified by these two methods.

PATIENTS AND METHODS During 1986 and 1987, a survey was conducted to determine the prevalence, detection, and treatment of alcoholism at The Johns Hopkins Hospital, a large urban hospital located in East Baltimore, Maryland. This research was designed as an observational period-prevalence study with approval of the Joint Committee on Clinical Investigation of The Johns Hopkins University School of Medicine. Two standardized alcoholism screening instruments, the CAGE and the SMAST [18-201, were used to determine the prevalence of screen-positive alcohol abuse, by department. These prevalences were as follows: Medicine (25%), Psychiatry (30%), Neurology (19%), Obstetrics/Gynecology (12.5%), and Surgery (23%). Detection rates by the housestaff and faculty physicians caring for those patients with a positive alcohol screen were less than 25% in Surgery and Obstetrics/Gynecology, between 25% and 50% in Neurology and Medicine, and greater than 50% in Psychiatry. This study was reported elsewhere [27]. Patients admitted to these adult clinical departments were surveyed regarding their alcohol consumption and other habits by trained interviewers who administered a standardized questionnaire. Informed consent was obtained from all patients enrolled in the study. Alcohol use was not the primary reason for obtaining consent, since information on other lifestyle habits was also obtained. Each patient completed a questionnaire requesting information on demographic characteristics, socioeconomic status, lifestyle habits, and questions on alcohol consumption as identified by the four-item CAGE questionnaire (Have you ever felt you should Cut down on your drinking? Have people Annoyed you by criticizing your drinking? Have you ever felt Guilty about your drinking? Have you ever used a drink in the morning as an Eye opener? [16,17]) and

MORBIDITY

/ UMBRICHT-SCHNEITER

ET AL

the Short Michigan Alcoholism Screening Test (SMAST, a questionnaire that assesses adverse psychosocial consequences of alcohol abuse, [20]). Patients admitted to each department were identified within 48 hours of their admission to the hospital for enrollment into the study. Patients were excluded if they were admitted to a short-stay unit (24 to 48 hours) for elective procedures. Identified patients who were too ill to participate were approached at a later time during their hospitalization. Of the 2,534 newly admitted patients, 532 were either too ill or refused to participate, which provided a total of 2,002 patients (79% response rate) who were interviewed for the study. We have no further information regarding characteristics of patients who declined to participate; however, the refusal rate was similar among the departments. Discharge diagnostic codes, based upon ICD9-CM criteria [13], were obtained for 1,964 (98%) of the 2,002 patients enrolled in the original study. Selected ICD-9-CM codes were stratified into diagnostic categories for our analyses (Appendix A). Alcohol-related discharge diagnostic codes were also selected in accordance with the classification used by the NIAAA in the Alcohol Epidemiologic Data System [12]. These alcoholrelated diagnoses are alcoholic psychosis, alcohol-dependency syndrome (including alcoholic polyneuropathy, alcoholic cardiomyopathy, and alcoholic gastritis), chronic liver diseases (cirrhosis), and nondependent alcohol abuse (Appendix B). Patients were defined as abusing alcohol when they either had an alcohol-related discharge diagnosis in their medical records or acknowledged experiencing adverse psychosocial consequences due to their alcohol consumption, as suggested by a positive alcoholism screen. There is no absolute gold standard for a diagnosis of alcohol abuse or dependence, but the most widely recognized instrument is a structured interview using DSM-III-R criteria for alcohol abuse or dependence [28]. We did not attempt to distinguish between alcohol abuse or dependence in this study. Patients were stratified into four groups: those with an alcohol-related diagnosis assigned as the principal discharge diagnosis (Group 1); those with an alcohol-related diagnosis assigned as a secondary but not principal discharge diagnosis (Group 2); those who screened positive for alcohol abuse but whose medical record indicated no alcoholrelated discharge diagnosis (Group 3; a positive screen was determined by two or more affirmative responses to the CAGE questionnaire and/or by a score of five or more points on the SMAST); and a reference group composed of patients who screened

August

1991

The American

Journal

of Medicine

Volume

91

111

ALCOHOL-ASSOCIATED MORBIDITY / UMBRICHT-SCHNEITER ET AL

Medicine

Ob-Gyn

Psychiatry

Neurology

Total () % patients

n 0

q

Group Group Group

10

1: Alcohol-Related 2: Alcohol-Related 3: Positive Screen

diagnosis as the Principal Discharge diagnosis as s Secondary Discharge for alcohol abuse, no Alcohol-Related

Diagnosis Diagnosis Diagnosis

Figure alcohol

negativefor alcoholabuseand did not havean alcohol-related dischargediagnosis recorded on their medical record. Information, from a random sample of patients, on method of payment and family income was retrieved from the original study data [27]. The risk for other comorbid conditionswascalculated by using the odds ratio (and corresponding 95%confidenceinterval) for other dischargediag-

1. Distribution abuse groups.

of the

three

noses between each of the three alcohol abuse groupsand the referencegroup.Analyseswereconducted for eachdepartment. Data from all departments were combined for some diagnoses,and, in orderto avoid a spuriouslyincreasedoddsratio, the denominator included the number of patients hospitalized in thosedepartmentswherethe diagnosis consideredwas cited at least once. Fisher’s exact test was used to determine statistical significance.

TABLE I Demographicand SocioeconomicVariables of Patients (%I in the Three Alcohol Abuse Groups Variable

Group2

(Gnro=y;)

(n = 117)

Group 3 (n = 295)

‘Z I:!; 7 (25)

iFI I2 2: I:?

‘ii I;;; Y I::’

83/117 (71)*

129/295 t441*

Reference (n = 1,524)

Age - (vears) .,~

17-29 30-49 50-69 270

Sex (male) Insurance Medicare Medicaid Blue Cross Private g$Y

$8

(68)*

l/28 (4)+ 14/28 (50)’ 7128 (25) 3128 (11) 3’2sor11’

Ei I;;; 330 (22) 45 (3) 404/1,521 (27)

20/113 (18) 48/l 13 (43)* 7/l 13 (6)* 10/113 (9) 25/l 13 (22)+ 3/l 13 (65)*

lncnmpt

94 (49) ? IF”

7 (9)’ 1 w+

:: I$

28 (15)

DIJP1 = alcohol-relateddiagnosisassigned as the principal discharge diagnosis; Group 2 = alcohol-related diagnosisassigned as a secondary discharge diagnosis; Group 3 = positive screen for alcuhl use, but no discharge diagnosis. ~0.01 awnpared with the referencegmup. t0.05compared with the reference group. icome status known anly in a sample of each group.

112

August 1991 The American Journal of Medicine

Volume 91

ALCOHOL-ASSOCIATED

Comparisons of demographic characteristics between the groups were done by Fisher’s exact test, the t-test, or Wilcoxon’s rank sum test, depending upon the variable.

MORBIDITY

/ UMBRICHT-SCHNEITER

ET AL

TABLE II Other Comorbid ConditionsAssociatedwith Group 1

RESULTS

Diagnosis

The distribution of patients across the four groups is shown in Figure 1 by department and for all departments combined. A principal discharge diagnosis of alcohol abuse was found in 1.4% of the patients (Group 1); 6% had an alcohol-related discharge diagnosis assigned as a secondary discharge diagnosis (Group 2); and 15% screened positive for alcohol abuse but no alcohol-related discharge diagnosis was recorded in their medical record (Group 3). Thus, the overall prevalence of alcohol abuse for the three diagnostic groups combined was 22.4%. Demographic and socioeconomic characteristics of the patients are shown in Table I for all departments combined. No age differences were found among the four groups. When compared with the reference group, a higher proportion of males was found in the three alcohol abuse groups. Patients in Groups 1 and 2 were more likely to be on Medicaid and have a lower income, findings consistent with a lower socioeconomic status. Patients in Group 3 had similar socioeconomic data when compared with the reference group. For patients in Group 1, the odds ratios and 95% confidence intervals for other comorbid conditions are shown in Table II. These patients were significantly more likely to have gastrointestinal diseases such as nonalcoholic liver disease, gastrointestinal bleeding, and pancreatitis. Other diagnoses included illegal drug abuse and chronic obstructive pulmonary disease (COPD). For patients in Group 2, the odds ratios and 95% confidence intervals for other comorbid conditions are shown in Table III. These patients had a significantly higher risk for cardiovascular disease including hypertension, as well as hypotension (orthostatic or chronic hypotension). Gastrointestinal diseases included pancreatitis (odds ratio, 24), nonalcoholic liver diseases (odds ratio, 7.5), gastrointestinal bleeding (odds ratio, 5.5), and peptic ulcer disease (odds ratio, 3.3). Psychiatric conditions included affective disorders, personality disorders, and poly-drug abuse, such as illegal drug abuse, nicotine dependence, and abuse of nonprescription drugs. Respiratory diseases included pneumonia and COPD. Patients in Group 2 were also at higher risk for trauma with a diagnosis of injury being six times more frequent when compared with the reference group. Acquired anemia, seizure disorder, and urinary tract infection were

Group 1

Reference

Odds Ratio

95% Confidence Interval

56.9*

17.0-190

Medicine Other liver disease Dementia Pancreatitis Substance abuse 1 1

Combined departmenkt Other liver disease Pancreatitis Substance abuse COPD,,asthma Gastromtestinal bleeding

6128 2127 3128 !E

6/1,524 9/612 32/l ,4 5511,5 2311,524

1.1-26.1

e-la.2 1.4-12.7 2.4+

1.1-21.5

J

Group 1 = alcohol-relateddiagnosis assignedas the principal discharge diagnaISIS. tp co.01 _ __ compared with the referencegroup. ‘p tU.U~compared with the referencegroup. tFor combined departments, denominators include only those departments in which the diagnosis occurred at least once.

other comorbid conditions found significantly more often in Group 2. For patients in Group 3, the odds ratios and 95% confidence intervals of other comorbid conditions are shown in Table IV. These patients were significantly more likely to have a diagnosis of injury, arrhythmia, hypertension, COPD, or asthma. Noncompliance with medication tended to be twice as prevalent (95% interval: 0.94 to 4.5; p = 0.10). Within individual departments, patients in Group 3 were at increased risk for breast cancer (Surgery), pelvic inflammatory disease (Obstetrics/Gynecology), and sexual deviation (Psychiatry).

COMMENTS Two of the studied alcohol abuse groups were defined by principal and secondary alcohol-related discharge diagnoses, using definitions developed by the NIAAA to study the epidemiology of alcoholrelated morbidity in the National Hospital Discharge Survey. In our hospital, the prevalence of alcohol-related discharge diagnoses is comparable to the data obtained in the National Hospital Discharge Survey [14]. These criteria alone, however, detected only one third of the hospitalized alcohol abusers detected using alcoholism screening questionnaires. The patients detected using the CAGE and the SMAST are a heterogeneous group in regard to their alcohol abuse pattern. They include patients with less severe or later onset alcoholism, episodic alcohol abusers, and recovering alcoholics. Questionnaires that focus on the adverse psychosocial consequences of alcohol use are effective tools for physicians in raising their suspicion of a patient’s alcohol abuse in its early stages. The CAGE and

August

1991

The American

Journal

of Medicine

Volume

91

113

ALCOHOL-ASSOCIATED MORBIDITY / UMBRICHT-SCHNEITER ET AL Ill Other Comorbid ConditionsAssociatedwith Group 2 TABLE

95% Diagnosis Medicine Pancreatitis Illegal substance abuse Pneumonia Noncompliance Seizure disorder Gastrointestinal bleeding COPO, asthma Acquired anemia

Surgery

Injury Pancreatitis Illegal substance abuse

Psychiatry Nonprescription substance abuse illegal substance abuse Combined departmen& Pancreatitis Illegal substance abuse Gastrointestinal bleeding Pneumonia Nonprescription substance abuse Injury Other liver disease Personality disorder Nicotine dependence Acquired anemia Hypotension COPD, asthma Peptic ulcer disease Seizure disorder Affective disorder Urinary tract infection

Group 2

Reference

%EIi

16158 10158

71281

19/58

:z 16/281 271281 251281

9/18

13/331

12158

2.9-59.8

4.2’ 2.6’

1.5-12.0 1.1-6.3

9/612 32/1,429

23.9* 10.7*

10.4-55.0 6.C-19.1 3.6-15.4

2%f4 1 l/503 3;;; ,;;‘:

2; g* 5:5+

2.5-18.8 2.6-20.2

:i;% 9611.524 N/l;524 5511,524 y&24

;.;:

lo/172 201172

‘G7 E241117

EY 1:2-5.2 1.1-4.0

13.0*

;;::

;;z;

8.4-82.7 3.8-35.2 2.6-20.2 2.1-14.7

8.3-71.9 4.5-77.0

z;%

:s;ii:

.-

5.6*

Confidence Interval

24.5* 18.6*

.%

20176

Odds Ratio

10111.524 .~~

90/l ,524 256/1,524

3:8* 3.5’

33:;: * E’ 2.7* 1.95*

1.1-27.6 3.2-12.3 2.1-14.1 1.5-13.8

2.3-6.3 1.3-9.7 1.8-6.4 1.5-6.9 1.1-6.7 1.6-4.7 1.6-4.7 1.3-3.0

oup 2 = alcohol-relateddiagnosisassigned as a secondary discharge diagnosis. < 0.05 compared with the reference group. < 0.01 mmpared with the reference group. or combineddepartments, denominators include only those departments in which the diagnosis occurred at least once.

SMAST are two DSM-III-R-validated alcoholism screeningtests shownto havea sensitivity typically between 80% and 90% and a specificity between 75% and 95% at the cutoff used in our study [l&19,29-32]. Given an estimated 20%prevalence of alcohol abuse,the positive predictive value of the CAGE questionnaire, using a cutoff scoreof two, hasbeenshownto be approximately 60%[181.Some patients, despite severe alcohol dependence,will deny any relation between their problems and alcohol abuse and have false-negativealcoholism screens.This is inherent to the problem of alcoholism, and it alsoexplainswhy the total prevalenceof alcoholism in this study (22.4%)is slightly higher than was found in our first study (21%),as a few patients with an alcohol-relateddischargediagnosis did not screenpositive. Another limitation of alcoholism screeninginstruments is that they elicit a lifetime prevalenceof alcoholabuseproblemsand

114

August 1991 The American Journal of Medlclne

do not allow for a distinction betweencurrent and past abuse.However,evena past history of alcohol abuseis important information for physicianssince it may alert them to a vulnerability toward other addictive substancesthat may be displayedby their patients. The literature is rich in documentation about the toxicity of alcohol on different organ systems [33-351,including cardiovascular[5,36-42],gastrointestinal [43], hematologic [44], neurologic [45], and pulmonary [46,47]diseases.The associationof alcohol abusewith malignancies [4850] and with psychiatric diagnoses[22]is equallywell documented. Our study wasnot designedto necessarilyidentify new diagnosesassociatedwith alcohol abuse but to better define the increasein risk for selected diagnosesin hospitalizedpatients who abusedalcohol, comparedwith the risk found in a similar population who did not abusealcohol. The high odds

Volume 91

ALCOHOL-ASSOCIATED

MORBIDITY

/ UMBRICHT-SCHNEITER

ET AL

TABLE IV Other Comorbid Conditions Associatedwith Group 3 95%

Diagnosis

Group3

Reference

16/58

%:!

Medicine

OddsRatio(p) 1.9 (0.08) 1.7 (0.10)

Arrhythmia Hypertension

26/58

Surgery Breast cancer Injury COPD, asthma

1 l/87 9187

13/331 1 l/33 1

5.4* 3.5* 3.4*

1.3-21.5 1.5-8.2

12/90

451645

2.1*

1s4.0

20/172

2.8*

1.2-6.4

Obstetrics/gynecology Pelvic inflammatory disease

5126

71158

Psychiatry Sexual deviation

l/41

0.98-3.7

0.95-3.01

1.3-8.4

Neurology Affective disorder Injury COPD, asthma Arrhythmia Hypertension Noncompliance oup 3 = positive screen for alcohol abuse, but no discharge diagnosis. , < 0.05 compared with the reference group. or combined deportments, denominators includeonly those departments in which the diagnosisoccurred at least once. tP co.01 compared with the reference group.

ratios found for somediagnosesmake it imperative for physiciansto seriouslyconsideralcoholabuseby patients with thesediagnoses.For example,the risk of pancreatitis is 24 times higher; the risk of using illegal drugs is 10times higher; the risk of noncompliancewith medication is almost nine times higher; and in Surgery,the risk of injury is 24.5times higher for patients with an alcohol-related discharge diagnosis. Very little quantitative information exists regarding the increasedrisk for other conditions in patients who have a positive screen for alcohol abusebut no alcohol-related diagnosis.Theoretically, such knowledgecould help physicianstarget screeningat the patient population presentingwith those conditions. We found that diagnosessuch as hypertension, arrhythmia, pneumonia, COPD, injury, breastcancer,pelvic inflammatory disease,affective disorder,or noncompliancewith medication should raise the physician’s suspicion of alcohol abusein hospitalizedpatients.Notably, screen-positive patients were found to havenearly a threefold increasedrelative risk for injury comparedwith the referencegroup,which supportsother data indicating that alcohol abuse is a major risk factor for trauma. A national surveyreportedthat only 55%of the major trauma centers in the United States screen their patients for alcohol abuse, and less

than one third of them use alcoholism counselors [51]. In our survey, there were 59 patients with a dischargediagnosisof injury, half of whom had an alcohol-relateddischargediagnosisbut onequarter of whom were undetected although they screened positive for alcohol abuse.With the exception of injury, however,the oddsratios for other comorbid conditions are not sufficiently high for the physician to rely on their presenceasspecific markersfor alcohol abuse.When present,however,these diagnosesshould alert physicians to the possibility of alcohol abuse. Routine screeningof patients for alcohol abuse has beenadvocated,basedon the assumption that effective,early intervention will give the patient an opportunity to modify his/her drinking habits before irreversible organic complications have developed. Such secondary prevention of alcoholism should becomea major objective of medical institutions. It is important that health careproviders address the possibility of alcohol abuse in their patients when coordinating their managementsince continued heavy drinking can result in seriousillnessand premature death. CONCLUSION Our study showsthat, in patients hospitalized in short-term acute-carehospitals, the prevalenceof

August 1991 The American Journal of Medicine

Volume 91

115

ALCOHOL-ASSOCIATED MORBIDITY / UMBRICHT-SCHNEITER ET AL

alcohol abuse identified by an alcohol-related diagnosis is substantially less when compared with the prevalence as determined by an alcoholism screening test. Patients with a secondary alcohol-related diagnosis were at a very high risk for pancreatitis, dementia, sequelae of liver disease, injury, illegal substance abuse, and noncompliance with medication, suggesting that alcohol abuse should be considered in these patients until proven otherwise, and that for epidemiologic research purposes, these diagnoses should be considered alcohol-related. Patients who screened positive for alcoholism but did not have an alcohol-related discharge diagnosis were also at increased risk for several comorbid con-

ditions. Diagnoses such as hypertension, arrhythmia, pneumonia, COPD, injury, breast cancer, pelvic inflammatory disease, affective disorder, or noncompliance with medication should raise the physician’s suspicion of alcohol abuse as a predisposing or contributory factor. Secondary prevention of alcohol abuse can only be achieved by routinely screening every patient using recognized alcoholism screening instruments.

ACKNOWLEDGMENT We are grateful to Dr. L. Randol Barker for his thoughtful and constructive review of this manuscript, and to Mr. Joel Gibson and Ms. Carol Lent for their assistance in the preparation of this manuscript.

APPENDIX A Other Comorbid Conditions:Abbreviations, Definitions, and ICD-9-M Codes Cardiovascular HTN: Hypotension: CAD, MI, s/p CABG: Pericarditis: Endocarditis: Myocarditis: Cardiomyopathy: Arrhythmia: CHF: Periph vast dis: Acute arterial occl: LE varices:

Essential hypertension, complications of HTN. 401,403,404. 458. Coronary artery disease, angina, chronic ischemic heart disease, unstable angina, SEMI, cardiac arteriosclerosis, myccardial infarction (acute or old), status post-coronary artery bypass graft. 411,412,413,414,429,V458.1. 420. Endocarditis, valve damage secondary to. . . 421,424. 422,429.0. Cardiomyopathy, alcoholic cardiomyopathy, cardiomegaly. 425,429.9,429.3. Arrhythmia, status post-pacemaker placement. 427, V458.9. Congestive heart failure. 402,428. Peripheral vascular disease, arteriosclerosis, aortic aneurysm, other aneurysm, Raynaud, Buerger fthromboangiitis obliterans). 440-443,447. Acute arterial occlusion fembolic and thrombotic). 444. Varicose veins of the legs. 454.

Gastrointestinal Esolh varices:

Ale liver disease:

Esophageal varices with and without bleed. 456. Peptic ulcer disease, esophagitis, stricture and stenosis, rupture, Mallory-Weiss, bleed, leukoplakia, gastritis, achlorhydria, persistent vomiting, dyspepsia, nonspecific gastritis, gastric and duodenal ulcer. 530-536. Functional disorders of intestine, irritable colon, functional diarrhea, atony of colon, nonspecific. 564.1,564.5,564.8,564.9. Alcoholic chronic liver disease, cirrhosis, alcoholic fatty liver, alcoholic hepatitis, alcoholic liver damage, chmnic hepatitis.

Other liver dis:

Other liver diseases, liver necrosis, hepatic coma, hepatorenal syndrome.

Fct dis intest:

Pancreatitis: GI bleed:

!i71. _. _.

570,572.2-572.8,573,789.1,789.5. 577. Gastrointestinal

bleeding, hematemesis, melena, nonspecific etiology. 578.

Genitourinaty,renal Nephropathy: ARF: CRF etc.: UTI: BPH: PID: Vaginitis: Endometriosis: Ovarian c st: Abnorma Ymenses: Tubal infer-t: Fern gen pain: UT leiomyoma: Acquired anemia:

Glomerulonephritis, nephrotic syndrome, chronic glomerulonephritis, nephritis. 580-583. Acute renal failure. 584. Chronic renal failure, renal failure nonspecific, renal sclerosis, complications of renal failure. 585,586,5BB. Urinary tract infection, pyelonephritis, infectious cystitis, urethritis, hematuria. 595,597,599. Benign prostatic hyperplasia. 600. Pelvic inflammatory disease. 614. 616. 617. Ovarian cyst. 620. Abnormal menses, amenorrhea, polymenorrhea, etc. 626. Tubal infertility. 628. Pain originating in female genital organs. 625. Uterine leiomyoma. 218. 2ao-281,285.0,285.i, 285.9 Continued on next page

116

August 1991 The American Journal of Medicine

Volume 91

ALCOHOL-ASSOCIATED MORBIDITY / UMBRICHT-SCHNEITER ET AL APPENDIX A Kont’d) Other Comorbid Conditions: Abbreviations, Definitions, and ICDd-CM Codes Metabolic Diabetes mellitus: Obesity: Dehydration:

250. Obesity. 278. Dehydration, disorder of water and electrolytes.

276.5,276.8

Nett;o~ system Cerebrovascular accident, subarachnoid hemorrhage, intracranial hemorrhage, SubduraVextradural hemorrhage, occlusion of precerebral arteries, occlusion of intracerebral arteries, transient ischemic attack, acute ill-defined cerebrovascular disease, late effects of CVA.

342,344,430-438,784.3,784.5.

Seizure disorder: Migraine: Mental retard: Dementia:

Convulsion, epilepsy. 780, 345. Migraine headache. 346. Mental retardation. 317-319. Dementia. 290,291.2,292.8,294.1

Psychiatric Affective disorder: Adjustment disorder: Neurotic disorder: Personality disorder: Psychosomat: Substance abuse:

Affective disorders. 296.311. 309. Neurotic disorder. 300. 301. Psychosomaticdisorders. 316,306. Illegal substance abuse or dependence.

Nicotine depend.: Non-prescription subst. ab.: Non-compliance: Sexual deviation: Sexual dysfunction:

Nicotine dependence. 305.1 Laxatives, over-the-counter drugs, nonsteroidal anti-inflammatory Non-compliance. V15.81.

304,305.2-305.91.

agents, etc.

.). 305.9.

302.0-302.6,302.8-302.9. 302.7.

Respiratory Pneumonia:

Pneumonia (viral, pneumocococcal,

other bacterial, aspiration pneumonia, pneumonitis secondary to food or vomiting). 480-

483,485,486,507.

COPD, asthma: Pleural eff.: Lung abscess: Acute pulm edema:

Chronic obstructive pulmonary disease, chronic bronchitis, emphysema, asthma. 4911193,496. Pleural effusion, empyema, pleurisy. 510, 511.1, 511.2. 513.0,513.1. Acute pulmonary edema (excluded from CHF), respiratory failure after procedure or trauma. 518.4, 518.5.

H/o surgery: Injury: Malunion: Mech device complic: Post-op complication:

Prior history of surgery. V15.2. Injury. _^^ ^ 800-904,910-959.

ltwi

Mechanical complication of an implanted device. 996. Postoperative complications. 909.3,997,998,429,564,569,576, E780-E799. Neoplasms with distant metastasis and/or adenopathy(ies).

196-199. Ma;f;n$ ENT CA: Esoph CA: Gastric CA: Hepatic CA: Lung CA: Pancreas CA: Breast CA: Gyn CA: Rentbladder CA: Hematol malig:

Colon cancer. V10.05. Cancer of the mouth, nose, and throat area. 144, v10.01, v10.02. Cancerof the esophagus. 150, V10.03. Gastriccancer. 151. V10.04. Cancer of the liver. i55, V10.07. Cancer of the lung. 162, V10.11. Cancer of the pancreas. 157. Cancer of the breast. 174, V10.3. Gvnecoloeiccancer. 179-184. V10.4. dncer &he kidney, bladder, and urothelium. 188, V10.5. Hematologic malignancy. 200-208.

Complicationsoi pregnancy Ectopic preg: Abortion: Drug dependence: Eclampsia: Fetal defect: Placenta praevia:

Ectopic pregnancy. 633. Spontaneous abortion, legally induced abortion, illegally induced abortion, unspecified abortion. 634637. Drug dependence in the mother. 648. Eclampsia, pre-eclampsia, hypertension of pregnancy. 642. Known or suspected fetal abnormality affecting the management of the mother. 655. ;l$enta praevla, antepartum hemorrhage, abrupbo placenta.

Early labor:

Threatened labor, premature labor. 644.

August 1991 The American Journal of Medicine

Voluve 91

117

ALCOHOL-ASSOCIATED

MORBIDITV / UMBRICHT-SCHNEITERET AL

APPENDIX B Alcohol-Related Diagnoses(ICD-9-W* Akohdic psychosis Alcohol withdrawal delirium, alcohol amnestic syndrome, other alcoholic dementia, alcohol withdrawal hallucinosis, idiosyncratic alcohol intoxication, alcoholic jealousy, other specified alcoholic psychosis, unspecified alcoholic psychosis. (291.0-291.9) Alcohd dependence syndrome Pellagra. (265.2) Acute alcoholic intoxication (in alcoholism). (303.0) Other and unspecified alcohol dependence. (303.9) Alcoholic polyneuropathy. (357.5) Alcoholic cardiomyopathy. (425.5) Alcoholic gastritis. (535.3) Chronic liver disease and cirrhosis Alcoholic fatty liver, acute alcoholic hepatitis, alcoholic cirrhosis of the liver, alcoholic liver damage, unspecified. (571.0-571.3) Chronic hepatitis. (571.4) Biliary cirrhosis. (571.6) Other chronic nonalcoholic liver disease. (571.8) Portal hypertension. (572.3) Cirrhosis of the liver without mention of alcohol. (571.5) Unspecified chronic liver disease. (571.9) Nondependent abuse ol alcohol Alcohol abuse. (305.0) rom the U.S. Alcohol EpidemiologicData ReferenceManual, section 4.

REFERENCES 1. Hurt RD, Finlayson RE. Morse RM, eta/. Alcoholism in elderly persons: medical aspects and prognosis of 216 inpatients. Mayo Clin Proc 1988; 63: 753-60. 2. Kolb D, Gunderson E. Medical histories of problem drinkers during their first twelve years of naval service. J Stud Alcohol 1983; 44: 84-94. 3. Schmidt W. Popham RE. The role of drinking and smoking in mortality from cancer and other causes in male alcoholics. Cancer 1981; 47: 1031-41. 4. Urban0 MA, Estruch R, Navarro LF, eta/. The effect of alcoholism on skeletal and cardiac muscle. N Engl J Med 1989; 320: 409-15. 5. Ashley MJ. Olin JS. LeRiche WH, eta/. The physical disease characteristics of inpatient alcoholics. J Stud Alcohol 1981; 42: 1-14. 6. Mendelson JH, Babor TF. Mello NK. et al. Alcoholism and prevalence of medical and psychiatric disorders. J Stud Alcohol 1986; 47: 361-6. 7. German E. Medical problems in chronic alcoholic men. J Chron Dis 1973; 26: 661-8. 6. Pell S, D’Alonzo CA. The prevalence of chronic disease among problem drinkers. Arch Environ Health 1968; 16: 679-84. 9. Klatsky AL, Friedman GD. Siegelaub AB. Alcohol and mortality: a ten-year Kaiser Permanente experience. Ann Intern Med 1981; 95: 139-45. 10. Schmidt W. DeLint J. Causes of death of alcoholics. Q J Stud Alcohol 1972; 33: 171-85. 11. Bowen OR, Sammons JH. The alcohol-abusing patient: a challenge to the profession. JAMA 1988; 260: 2267-70. 12. Colliver JD, Kern JC, Hassett C. A system to convert ICD diagnostic codes for alcohol research. MMWR 19% 16: 216-22. 13. World Health Organization. The international classification of diseases, 9th revision, clinical modiiication. Ann Arbor: Commission on Professional and Hospital Activities, 1980. 14. Stinson FS, Aitken SS, Noble J. Hospital discharges with alcohol-related conditions: hospital discharge survey 1979-85. US. alcohol epidemiologic data reference manual, ~014, ADM 281-82-ooO3. Washington, DC: US Government Printing Office, 1989. 15. Williams GD, Grant BF, Stinson FS, eta/. Trends in alcohol-related morbidity and mortality. Public Health Rep 1988; 103: 592-7. 16. Allen JP. Eckardt MJ. Wallen J. Screening for alcoholism: techniques and issues. Public Health Rep 198.8; 103: 586-92. 17. Seizer ML. The Michigan Alcoholism Screening Test: the quest for a new diagnostic instrument. Am J Psychiatry 1971; 127: 1653-8. 18. Bush B. Shaw S. Cleary P. eta/. Screening for alcohol abuse using the CAGE questionnaire. Am J Med 1987; 82: 231-5.

118

August 1991 The American Journal of Medicine

19. Ewing JA. Detecting alcoholism: the CAGE questionnaire. JAMA 1984; 252: 1905-7. 20. Seizer ML. Vinokur A, VanRooljen L. A self-administered Short Michigan Alcoholism Screening Test (SMAST). J Stud Alcohol 1975; 36: 117-26. 21. Swenson WM. Morse RM. The use of a self-administered alcoholism screening test (SAAST) in a medical center. Mayo Clin Proc 1975; 50: 204-8. 22. Curtis JL. Millman EJ. Joseph M, et al. Prevalence rates for alcoholism, associated depression and dementia on the Harlem Hospital Medicine and Surgery Services. Adv Alcohol Subst Abuse 1986: 6: 4564. 23. Chick J. Lloyd GG. Duffy JC, eta/. Medical admissions in men: the risk among drinkers. Lancet 1986; 2: 13X1-2. 24. Lloyd G. Chick J, Crombie E. et al. Problem drinkers in medical wards: consumption patterns and disabilities in newly identified male cases. Br J Addict 1986; 81: 789-95. 25. Gomberg ES. Prevalence of alcoholism among ward patlents in a Veterans Administration hospital. J Stud Alcohol 1975; 36: 1458-67. 26. Cleary PD. Miller M, Bush BT, et al. Prevalence and recognition of alcohol abuse in a primary care population. Am J Med 1988; 85: 466-71. 27. Moore RD, Bone LR, Geller G, eta/. Prevalence, detection, and treatment of alcoholism in hospitalized patients. JAMA 1989; 261: 403-7. 26. American Psychiatric Association. Diagnostic and statistical manual of mental disorders, 3rd ed, rev (DSM-III-R). Washington, DC: American Psychiatric Association, 1987. 29. Mayefield D, McLeod G, Hall P. The CAGE questionnaire: validation of a new alcoholism screening instrument. Am J Psychiatry 1974; 131: 1121-3. 30. Yersin B, Trisconi Y, Paccaud F, eta/. Accuracy of the Michigan Alcoholism Screening Test for screening of alcoholism in patients of a medical department. Arch Intern Med 1989; 149: 2071-4. 31. Garzotto N, Baratta S, Faccincani C. Validation of a screening questionnaire for alcoholism (MAST) in an Italian sample. Compr Psychiatry 1988; 29: 323-9. 32. Zung BJ. Reliability and validity of the short MAST among psychiatric inpatients. J Clin Psycho1 1984; 40: 347-50. 33. Eckardt MJ, Harford TC. Kaelber CT, et al. Health hazards associated with alcohol consumption. JAMA 1981; 246: 648-63. 34. Secretary of Health and Human Services. Sixth special report to the U.S. Congress on Alcohol and Health, DHHS publication no. 87-1519. Washington, DC: US Government Printing Office, 1987. 35. Lieber CS. Medical disorders of alcoholism. Baltimore: WB Saunders, 1982. 36. Diamond I. Alcoholic myopathy and cardiomyopathy [editorial]. N Engl J Med 1989; 320: 458-9. 37. Friedman GD. Klatsy AL, Siegelaub AB. Alcohol intake and hypertension. Ann Intern Med 1983; 98: 846-9. 36. Friedman HS. Vasavada BC, Malec AM, et al. Cardiac function in alcoholassociated systemic hypertension. Am J Cardiol 1986; 57: 227-31. 39. Gordon T. Kannel WB. Drinking and its relation to smoking, blood pressure, lipids, and uric acid. The Framingham Study. Arch Intern Med 1983; 143: 1366-74. 40. Klatsky AL, Friedman GD, Armstrong MA. The relationships between alcoholic beverage use and other traits to blood pressure: a new Kaiser Permanente study. Circulation 1986; 73: 628-36. 41. Moore RD. Alcohol and hypertension. Md Med J 1988; 37: 385-9. 42. Knochel JP. Cardiovascular effects of alcohol. Ann Intern Med 1983; 98: 849-54. 43. Lieber CS. Biochemical and molecular basis of alcohol-induced injury to liver and other tissues. N Engl J Med 1988: 319: 1639-50. 44. Colman N. Herbert V. Hematologic complications of alcoholism: overview. Semin Hematol 1980; 17: 164-76. 45. Charness ME, Simon RP. Greenberg DA. Ethanol and the nervous system. N Engl J Med 1989; 321: 442-50. 46. Bomalaski JS. Phair JP. Alcohol, immunosuppression, and the lung. Arch Intern Med 1982; 142: 2073-4. 47. Heinemann HO. Alcohol and the lung. Am J Med 1977; 63: 81-5. 48. Graham S. Alcohol and breast cancer [editorial]. N Engl J Med 1987; 316: 1211-2. 49. Schatzkin A, Jones Y, Hoover RN, et al. Alcohol consumption and breast cancer in the epidemiologic follow-up study of the first National Health and Nutrition Examination Survey. N Engl J Med 1987; 316: 1169-73. 50. Willett WC, Stamper MJ. Colditz GA, et al. Moderate alcohol consumption and the risk of breast cancer. N Engl J Med 1987; 316: 1174-W 51. Soderstrom CA, Cowley RA. A national alcohol and trauma center survey: missed opportunities, failures of responsibility. Arch Surg 1987; 122: 1067-71.

Volume 91