Evidence from anorexia nervosa at least consistent with CR being good

[Guest guest]

Hi All, Can you get http://www.mayo.edu/comm/mcr/news_1393.html = Mayo

Clinic healthy food with search option for the below paper? It is:

Mayo Clin Proc 2003 Mar;78(3):278-84

PMID: 12630579 [PubMed - indexed for MEDLINE].

The full PDF is available from email: apater@... in case you cannot

access it.

Evidence from anorexia nervosa at least consistent with CR being good was

presented to my mind.

Cheers, Al.

http://www.mayo.edu/comm/mcr/news_1393.html = Mayo Clinic healthy food with

search option for the below paper.

Mayo Clin Proc 2003 Mar;78(3):278-84

Comment in: Mayo Clin Proc. 2003 Mar;78(3):273-4.

Long-term survival of patients with anorexia nervosa: a population-based study

in Rochester, Minn.

Korndorfer SR, Lucas AR, Suman VJ, Crowson CS, Krahn LE, Melton LJ

OBJECTIVE: To estimate long-term survival of unselected patients with anorexia

nervosa from Rochester, Minn. PATIENTS AND METHODS: In this population-based

retrospective cohort study, all 208 Rochester residents who presented with

anorexia nervosa (193 women and 15 men) for the first time from 1935 through

1989 were monitored for up to 63 years. Subsequent survival was compared with

that expected for Minnesota white residents of similar age and sex, and

standardized mortality ratios were determined on the basis of age- and

sex-specific death rates for the US population in 1987. RESULTS: Survival was

not worse than expected in this cohort (P = .16). The estimated survival 30

years after the initial diagnosis of anorexia nervosa was 93% (95% confidence

interval, 88%-97%) compared with an expected 94%. During 5646 person-years of

follow-up (median, 22 years per patient), 17 deaths occurred (14 women and 3

men) compared with an expected 23.7 deaths (standardized mortality ratio, 0.71;

95% confidence interval, 0.42-1.09). One woman died of complications of anorexia

nervosa, 2 women committed suicide, and 6 patients (5 women and 1 man) died of

complications of alcoholism. Other causes of death were not increased.

CONCLUSIONS: Long-term survival of Rochester patients with anorexia nervosa did

not differ from that expected. This finding suggests that overall mortality was

not increased among the spectrum of cases representative of the community.

PMID: 12630579 [PubMed - indexed for MEDLINE]

Outcomes among patients with anorexia nervosa are

extremely variable.Many patients recover fully,oth-

ers have recurrent episodes or have chronic disease,and

some die of complications of their disease.1(pp186-208),2 A

general belief is that the overall risk of death is increased

among these patients,but investigations that have ad-

dressed this issue have varied in study population,diagnos-

tic criteria,extent of follow-up,and period analyzed.3 Most

of the early investigations reported crude mortality (the

proportion of patients who died during follow-up)ranging

from 0%to 22%.4,5 One review indicated that the crude

mortality appeared to have declined over time,from about

10%in the 1950s to about 4%in the 1980s,6 with recent

rates of 2%to 6%.7-9 However,these calculations do not

account for variability in the duration of follow-up,and

they do not compare survival to that which would be ex-

pected for similar individuals without the disease.These

problems are overcome by estimating standardized mortal-

ity ratios (SMRs),which indicate that death rates among

patients with anorexia nervosa are 1.4 to 17 times higher

than expected.9-20 Although these figures suggest that an-

orexia nervosa has a substantial negative effect on survival,

most of the studies were performed at referral centers,

which tend to treat severely affected patients who have had

multiple treatments and relapses.

In contrast to the high mortality rates reported in clinical

studies,an average of 145 deaths annually (from more than

10 million deaths registered with the National Center for

Health Statistics for 1986-1990)were attributed to an-

orexia nervosa.21 Although undoubtedly an underestimate

because anorexia nervosa may not be noted as the immedi-

ate or underlying cause of death,this suggests that the death

rate might be much lower than generally surmised from

clinical studies.To obtain a more representative picture,we

examined the survival of a population-based cohort of resi-

dents of Rochester,Minn,who met diagnostic criteria for

anorexia nervosa during the 55-year period 1935 through

1989 and who were subsequently monitored for up to 63

years.

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

RESULTS

Altogether,208 Rochester residents (193 women and 15

men)first met the criteria for anorexia nervosa in the 55-

year period 1935 through 1989.The patients ranged in age

from 10 to 57 years,with a median age at diagnosis of 19

years (mean,21.5 years),which was similar in women and

men.All the patients were white,reflecting the racial com-

position of the community (99%white in 1970).Eighty-

two patients (39%)had definite anorexia nervosa,92

(44%)had probable anorexia,and 34 (16%)had possible

anorexia.Of the female incidence cases,80 (41%)were

definite cases of anorexia nervosa,84 (44%)were probable

cases,and 29 (15%)were possible cases.Among the male

incidence cases,2 (13%)were definite cases,8 (53%)were

probable cases,and 5 (33%)were possible cases.

After anorexia nervosa was initially recognized,the

study cohort was followed up for a total of 5646 person-

years.This represents 92%of the total duration of follow-

up that would have been possible if all surviving patients

had been observed until the termination of update activities

(6106 person-years).The mean duration of follow-up was

27.1 years (median,22 years;range,1 day to 63 years),and

96%of patients were followed up for at least 10 years.At

last contact,17 patients (14 women and 3 men)had died.

The expected number of deaths in the age-and sex-

matched reference cohort of Minnesota white people was

23.7 (19.3 for women and 4.4 for men)for an overall SMR

of 0.71 (95%CI,0.42-1.09).Thus,overall survival among

the patients with anorexia nervosa was not worse than that

expected (P =.16)(Figure 1)

Figure 1.Observed and expected survival among Rochester,

Minn,residents first diagnosed with anorexia nervosa from 1935

through 1989.

>>>[There were fewer deaths from 40-60 weeks and then they were the same.

It suggests to me that fewer died when they had less weight and recovered

patients then died at normal or above rate.]

and did not differ between the

patients diagnosed in the first 30 years of the study and

those diagnosed in the last 25 years (P =.15).By 30 years

after the initial diagnosis,an estimated 93%(95%CI,88%-

97%)of the patients were still alive compared to an ex-

pected 94%.The SMR for this more restricted period of

observation was 1.30 (95%CI,0.62-2.38).Six of the

deaths occurred among the 82 patients (7%)with definite

anorexia nervosa,8 among the 92 patients (9%)with prob-

able anorexia nervosa,and 3 among the 34 patients (9%)

with possible anorexia nervosa.Survival did not differ

among these groups (P =.88),and the SMRs were 0.92

(95%CI,0.30-2.14),0.87 (95%CI,0.40-1.65),and 0.38

(95%CI,0.08-1.10),respectively.

The causes of death are shown in Table 1.

Table 1.Deaths Among Rochester,Minn,Residents First Diagnosed With Anorexia

Nervosa,1935-1989*

Age at

Age at Years

Underlying Other important

Year of Dx Weight Diagnostic

Clinical death after Immediate cause cause

of death underlying

Dx (y) recovered certainty

condition (y) Dx of death

condition

Women

1950 35 Yes Probable

Alcoholism 37 2 Hepatic failure

Alcoholic cirrhosis None

1960 27 No Definite Inanition 41 13 Cardiac arrest Anorexia nervosa None

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

*COPD =chronic obstructive pulmonary disease; NOS =not otherwise specified.

Of note,one

woman died of cardiac arrest at age 41 years (13 years after

her initial diagnosis),and anorexia nervosa was listed on

the death certificate as the underlying cause.This patient

had never sought treatment.When she was brought to the

emergency department,she was moribund and in a state of

severe inanition.She had a cardiac arrest and died shortly

after being admitted to the hospital.Anorexia nervosa was

not listed on the death certificates for the other patients.

However,2 women committed suicide (carbon monoxide

poisoning in each instance)at ages 33 and 47 years;this

was 14 and 17 years,respectively,after the initial diagnosis

of anorexia nervosa.Five women (ages 27,33,37,70,and

71 years)and 1 man (age 48 years)died of complications of

alcoholism 15,14,2,51,13,and 22 years,respectively,

after diagnosis.Other than for mental disorders,the distri-

bution of underlying causes of death did not differ signifi-

cantly from that expected (Table 2).

Table 2.Number of Deaths Observed and Expected and Standardized Mortality Ratio

by Underlying Cause Among Rochester,Minn,Residents First Diagnosed

With Anorexia Nervosa,1935-1989*

No.of deaths

Underlying cause Observed

Expected SMR 95%CI

Infectious and parasitic diseases 0 0.45

0.00 0.00-8.20

Neoplasms 3

7.75 0.39 0.08-1.13

Endocrine and metabolic diseases 0.70 0.00 0.00-5.26

Diseases of the blood 0 0.10 0.00 0.00-37.0

Mental disorders 2

0.22 9.14 1.11-32.9 †

Diseases of the nervous system 0 0.47 0.00 0.00-7.87

Diseases of the circulatory system 3 10.40

0.29 0.06-0.84

Diseases of the respiratory system 3 1.87

1.60 0.33-4.67

Diseases of the digestive system 2 1.03

1.95 0.24-7.02

Diseases of the genitourinary system 0 0.39 0.00 0.00-9.36

Complications of pregnancy 0 0.00 ……

Diseases of the skin 0 0.04 0.00 0.00-103

Diseases of the musculoskeletal system 0 0.12 0.00 0.00-31.7

Congenital and perinatal conditions 0 0.07 0.00 0.00-50.1

Ill-defined conditions 1

0.32 3.08 0.08-17.1

Injury and poisoning 3

2.21 1.36 0.28-3.96

Total

17 26.20

*CI =confidence interval; SMR =standardized mortality ratio.

†Statistically significant increase,P <.05.

DISCUSSION

In this first population-based study of unselected cases of

anorexia nervosa from Rochester,we found no reduction in

long-term survival compared with that expected for Minne-

sota white women and men of comparable age.Only 1

patient died of complications directly related to anorexia

nervosa.This relatively benign prognosis compared to pre-

vious reports is probably primarily due to the mild clinical

spectrum represented by unselected patients from the gen-

eral population.Indeed,most of the Rochester patients had

never seen a psychiatrist for anorexia nervosa,and few had

been hospitalized for the condition.24 However,all the pa-

tients were identified on the basis of signs,symptoms,or

diagnoses consistent with anorexia nervosa,and care was

taken to use the standard DSM-III-R criteria 25 extant at the

time of the study as well as the Pathology of Eating Group

criteria.26 The resulting incidence rates for anorexia

nervosa,24,36 which have been widely accepted as the stan-

dard in the field,are in close agreement with the best

studies in other countries.1(pp59-76),37,38 This suggests that the

diagnostic criteria used herein were not excessively broad

and that they resulted in a number of community cases

similar to those reported by others.

However,there were 2 suicides and 6 deaths related to

alcoholism;thus,all the underlying psychiatric problems

may not have been resolved.2 An association of anorexia

nervosa with depression has long been recognized,39-41 and

an increase in deaths due to suicide has been reported

previously.17,19,42,43 In contrast,data from the National Cen-

ter for Health Statistics suggest that suicide rates are not

increased among women when anorexia nervosa is listed

on the death certificate as an underlying cause or accompa-

nying cause of death.44 Although clinical studies are

skewed in the direction of greater pathology,death certifi-

cates likely underestimate the frequency of anorexia

nervosa as a cause of death and as a contributing cause of

suicide.Indeed,anorexia nervosa was not mentioned on the

death certificate of 16 of our 17 patients who died.

Deaths due to alcoholism appear to be common among

patients with eating disorders 19,45 but have received little

detailed attention 46-48 despite the fact that alcohol depen-

dence is overrepresented among patients with anorexia

nervosa.49 Conversely,Peveler and Fairburn 50 reported that

19%of women attending an alcohol treatment program had

a history of anorexia nervosa.Also,first-degree relatives of

patients with anorexia nervosa have significantly more al-

coholism than do relatives of controls,suggesting that there

may be a similar predisposition to the 2 disorders.51 The

association of alcoholism with bulimia nervosa is particu-

larly strong.48 Patients with bulimia nervosa,compared

with those with anorexia nervosa,are prone to impulsive

behaviors,including experimentation with alcohol,and it is

well known that a significant proportion of patients with

anorexia nervosa develop bulimia nervosa as their illness

progresses.52 Therefore,it is likely that some of our patients

became bulimic and began abusing alcohol.However,we

do not have sufficient clinical information to confirm this

supposition because many of the patients were not seen in

treatment.

Most patients with anorexia nervosa do not use alcohol.

However,some young women may secretly begin abusing

alcohol,become malnourished,and present with weight

loss that is diagnosed as anorexia nervosa.In at least one of

our patients,alcohol abuse occurred before she fulfilled the

diagnostic criteria for anorexia nervosa,but the alcohol

abuse was not known until years later.In addition,some

deaths due to pneumonia are attributable to alcoholism

among patients with psychiatric problems.53 Of impor-

tance,the 6 deaths due to alcoholism and/or pneumonia,

which might directly or indirectly relate to anorexia

nervosa,occurred a median of 13.5 years after the initial

diagnosis of anorexia nervosa.This suggests that a thor-

ough elicitation of history is necessary in patients with

anorexia nervosa to identify alcohol abuse.Additionally,

clinicians should monitor drinking behavior in such pa-

tients and provide close follow-up to ensure that this type

of behavior is identified and treated.In a similar vein,we

previously documented an increase in the risk of os-

teoporotic fractures among these patients,but the problem

did not become evident for decades after the diagnosis of

anorexia nervosa had first been made.54

Several previous reports have also compared observed

with expected survival,but the results show considerable

variation that is attributable to selection of the study popula-

tion.To our knowledge,no previous study has been strictly

population based.Perhaps closest to this ideal was a study of

patients with anorexia nervosa enumerated on the Aberdeen

case register.11 The SMR for that group was 4.7,which was

marginally significant given the 8 deaths observed;however,

more than half of the patients had required hospitalization,

and 20%had undergone electroconvulsive therapy.Conse-

quently,these patients seem to have a more serious spectrum

of disease compared with our community patients.Two

studies of patients in specialized clinics reported an SMR of

14,12,13 whereas SMRs for patients requiring hospitalization

have varied from 1.4 to 17.9-11,14-16,18-20 These findings indicate

that no one “best ” estimate of excess mortality exists and that

prognosis relates to the clinical spectrum of disease in the

patients being studied.

The major strengths of our study are the inclusion of a

representative sample of patients with anorexia nervosa

from the general population and the long-term follow-up

that is possible because of the unique medical records

system in Rochester.Corresponding limitations relate to

the relatively small number of patients diagnosed earlier in

the study period and the restriction of the study to white

patients due to the demographic composition of the com-

munity.However,the socioeconomic characteristics of the

local population resemble those of the US white population

in general.22 Moreover,the incidence of anorexia nervosa

in our population 36 is comparable to that in other recent

population-based studies.55,56 Follow-up was relatively

complete in our cohort (96%of subjects were followed up

for at least 10 years),but the number of deaths was low.

The limited numbers precluded a definitive exploration of

the distribution of causes of death,particularly in the men.

In addition,our data were insufficient to define precisely

the subgroups of patients who might have had an increased

risk of death.Specifically,we saw no difference in survival

by diagnostic certainty (definite,probable,or possible an-

orexia nervosa),but diagnostic certainty is not synonymous

with disease severity,although the two tend to be corre-

lated in practice.

Anorexia nervosa is a serious,often recurrent,and po-

tentially fatal illness that has been increasing in frequency

among young females.36 Although previous studies suggest

that a diagnosis of anorexia nervosa implies a severe nega-

tive effect on survival,we found that survival in our popu-

lation-based cohort of patients did not differ from that

expected.This probably relates to the relatively mild dis-

ease expressed by these unselected community patients,

few of whom were hospitalized or treated in a specialized

outpatient center and most of whom eventually recov-

ered.24 Prognosis is clearly worse among patients referred

to tertiary centers for treatment of anorexia nervosa.Al-

though our data suggest that overall mortality is not in-

creased among community patients with anorexia nervosa

in general,these findings should not lead to complacency

in clinical practice because deaths do occur.In addition,

more research is needed to define the association of suicide

and alcoholism in patients with anorexia nervosa.Early

recognition of anorexia nervosa and its appropriate treat-

ment are warranted.This requires a sensitive balance be-

tween empathic support and the expectation of restoring

normal eating patterns and weight.Efforts to enforce

weight gain that are too aggressive can reinforce patient

resistance and lead to chronicity.Associated psychological

issues also need to be addressed.

REFERENCES

1.Hsu LKG.Eating Disorders.New York,NY:Guilford Press;1990:

59-76,186-208.

2.Strober M,Freeman R,Morrell W.The long-term course of severe

anorexia nervosa in adolescents:survival analysis of recovery,

relapse,and outcome predictors over 10-15 years in a prospective

study.Int J Eat Disord.1997;22:339-360.

3.Ratnasuriya RH,Eisler I,Szmukler GI, GF.Anorexia

nervosa:outcome and prognostic factors after 20 years.Br J Psy-

chiatry.1991;158:495-502.

4.Hsu LK.Outcome of anorexia nervosa:a review of the literature

(1954 to 1978).Arch Gen Psychiatry.1980;37:1041-1046.

5.Herzog DB,Keller MB,Lavori PW.Outcome in anorexia nervosa

and bulimia nervosa:a review of the literature.J Nerv Ment Dis.

1988;176:131-143.

6.Steinhausen HC,Rauss-Mason C,Seidel R.Follow-up studies of

anorexia nervosa:a review of four decades of outcome research.

Psychol Med.1991;21:447-454.

7.Pla C,Toro J.Anorexia nervosa in a Spanish adolescent sample:an

8-year longitudinal study.Acta Psychiatr Scand.1999;100:441-

446.

8.Ramsay R,Ward A,Treasure J, GF.Compulsory treatment

in anorexia nervosa:short-term benefits and long-term mortality.

Br J Psychiatry.1999;175:147-153.

9.Fichter MM,Quadflieg N.Six-year course and outcome of anorex-

ia nervosa.Int J Eat Disord .1999;26:359-385.

10.Patton GC.Mortality in eating disorders.Psychol Med.1988;18:

947-951.

11.Crisp AH,Callender JS,Halek C,Hsu LK.Long-term mortality in

anorexia nervosa:a 20-year follow-up of the St ’s and

Aberdeen cohorts.Br J Psychiatry.1992;161:104-107.

12.Norring CE,Sohlberg SS.Outcome,recovery,relapse and mortal-

ity across six years in patients with clinical eating disorders.Acta

Psychiatr Scand.1993;87:437-444.

13.Deter HC,Herzog W.Anorexia nervosa in a long-term perspective:

results of the Heidelberg-Mannheim study.Psychosom Med.

1994;56:20-27.

14.Eckert ED,Halmi KA,Marchi P,Grove W,Crosby R.Ten-year

follow-up of anorexia nervosa:clinical course and outcome.

Psychol Med.1995;25:143-156.

15.Møller-Madsen S,Nystrup J,Nielsen S.Mortality in anorexia

nervosa in Denmark during the period 1970-1987.Acta Psychiatr

Scand.1996;94:454-459.

16.Herzog W,Deter HC,Fiehn W,Petzold E.Medical findings and

predictors of long-term physical outcome in anorexia nervosa:a

prospective,12-year follow-up study.Psychol Med.1997;27:269-

279.

17.Nielsen S,Møller-Madsen S,Isager T,Jørgensen J,Pagsberg K,

Theander S.Standardized mortality in eating disorders —quantita-

tive summary of previously published and new evidence.J

Psychosom Res.1998;44:413-434.

18.Crow S,Praus B,Thuras P.Mortality from eating disorders —a 5-

to 10-year record linkage study.Int J Eat Disord.1999;26:97-101.

19.Herzog DB,Greenwood DN,Dorer DJ,et al.Mortality in eating

disorders:a descriptive study.Int J Eat Disord.2000;28:20-26.

20.Löwe B,Zipfel S,Buchholz C,Dupont Y,Reas DL,Herzog W.

Long-term outcome of anorexia nervosa in a prospective 21-year

follow-up study.Psychol Med .2001;31:881-890.

21.Hewitt PL,Coren S,Steel GD.Death from anorexia nervosa:age

span and sex differences.Aging Ment Health .2001;5:41-46.

22.Melton LJ III.History of the Rochester Epidemiology Project.

Mayo Clin Proc.1996;71:266-274.

23.Kurland LT,Molgaard CA.The patient record in epidemiology.Sci

Am.1981;245:54-63.

24.Lucas AR,Beard CM,O ’Fallon WM,Kurland LT.Anorexia

nervosa in Rochester,Minnesota:a 45-year study.Mayo Clin Proc.

1988;63:433-442.

25.American Psychiatric Association.Diagnostic and Statistical

Manual of Mental Disorders .Revised 3rd ed.Washington,DC:

American Psychiatric Association;1987:65-67.

26.Garrow JS,Crisp AH,Jordan HA,et al.Pathology of Eating Group

report.In:Silverstone T,ed.Appetite and Food Intake:Report of

the Dahlem Workshop on Appetite and Food Intake,Berlin,1975,

December 8-12 .Berlin,Germany:Abakon-Verlagsgesellschaft;

1976:405-416.

27.Melton LJ III.The threat to medical-records research.N Engl J

Med.1997;337:1466-1470.

28.International Classification of Diseases,9th Revision:Clinical

Modification (ICD-9-CM).2nd ed.Washington,DC:US Dept of

Health and Human Services,Public Health Service,Health Care

Financing Administration;1980.DHHS publication PHS 80-1260.

29.Kaplan EL,Meier P.Nonparametric estimation from incomplete

observations.J Am Stat Assoc .1958;53:457-481.

30.Hakulinen T.Cancer survival corrected for heterogeneity in patient

withdrawal.Biometrics.1982;38:933-942.

31.Hakulinen T,Abeywickrama KH.A computer program package

for relative survival analysis.Comput Programs Biomed.1985;19:

197-207.

32.Fleming TR,Harrington DP.A class of hypothesis tests for one and

two sample censored survival data.Commun Stat Theor Meth.

1981;A10(8):763-794.

33.Bergstralh EJ,Offord KP,Kosanke JL,Augustine GA.A SAS

Procedure for Person Year Analyses .Rochester,Minn:Depart-

ment of Health Sciences Research;1986.Technical Report Series

No.31.

34.National Center for Health Statistics.Health US,1992 .Hyattsville,

Md:US Dept of Health and Human Services;1993.Publication

PHS 93-1232.

35.Kupper LL,Mc AJ,Symons MJ,Most BM.On the util-

ity of proportional mortality analysis.J Chronic Dis.1978;31:15-

22.

36.Lucas AR,Crowson CS,O ’Fallon WM,Melton LJ III.The ups and

downs of anorexia nervosa.Int J Eat Disord.1999;26:397-405.

37.van Hoeken D,Lucas AR,Hoek HW.Epidemiology.In:Hoek HW,

Treasure JL,Katzman MA,eds.Neurobiology in the Treatment of

Eating Disorders.Chichester,England: Wiley & Sons;1988:

97-126.

38.van ’t Hof S.Anorexia Nervosa:The Historical and Cultural Speci-

ficity:Fallacious Theories and Tenacious ‘Facts.’ Amsterdam,the

Netherlands:Swets & Zeitlinger;1994.

39.Hendren RL.Depression in anorexia nervosa.J Am Acad Child

Psychiatry .1983;22:59-62.

40.Swift WJ,s D,Barklage NE.The relationship between

affective disorder and eating disorders:a review of the literature.

Am J Psychiatry .1986;143:290-299.

41.Casper RC.Depression and eating disorders.Depress Anxiety .

1998;8(suppl 1):96-104.

42.Theander S.Outcome and prognosis in anorexia nervosa and bu-

limia:some results of previous investigations,compared with those

of a Swedish long-term study.J Psychiatr Res.1985;19:493-508.

43.Sullivan PF.Mortality in anorexia nervosa.Am J Psychiatry.1995;

152:1073-1074.

44.Coren S,Hewitt PL.Is anorexia nervosa associated with elevated

rates of suicide?Am J Public Health .1998;88:1206-1207.

45.Emborg C.Mortality and causes of death in eating disorders in

Denmark 1970-1993:a case register study.Int J Eat Disord.1999;

25:243-251.

46.Goldbloom DS.Alcohol misuse and eating disorders:aspects of an

association.Alcohol Alcohol .1993;28:375-381.

47.Holderness CC,-Gunn J,Warren MP.Co-morbidity of

eating disorders and substance abuse review of the literature.Int J

Eat Disord .1994;16:1-34.

48.Schuckit MA,Tipp JE,Anthenelli RM,Bucholz KK,Hesselbrock

VM,Nurnberger JI Jr.Anorexia nervosa and bulimia nervosa in

alcohol-dependent men and women and their relatives.Am J Psy-

chiatry .1996;153:74-82.

49.Sullivan PF,Bulik CM,Fear JL,Pickering A.Outcome of anorexia

nervosa:a case-control study.Am J Psychiatry.1998;155:939-946.

50.Peveler R,Fairburn C.Eating disorders in women who abuse

alcohol.Br J Addict.1990;85:1633-1638.

51.Halmi KA,Eckert E,Marchi P,Sampugnaro V,Apple R,Cohen J.

Comorbidity of psychiatric diagnoses in anorexia nervosa.Arch

Gen Psychiatry.1991;48:712-718.

52. G.Bulimia nervosa:an ominous variant of anorexia

nervosa.Psychol Med.1979;9:429-448.

53.Zilber N,Schufman N,Lerner Y.Mortality among psychiatric

patients —the groups at risk.Acta Psychiatr Scand.1989;79:248-

256.

54.Lucas AR,Melton LJ III,Crowson CS,O ’Fallon WM.Long-term

fracture risk among women with anorexia nervosa:a population-

based cohort study.Mayo Clin Proc.1999;74:972-977.

55.Møller-Madsen SM,Nystrup J.Increased incidence of anorexia

nervosa in Denmark [in Danish ].Ugeskr Laeger .1994;156:3291-

3293.

56.Hoek HW,Bartelds AI,Bosveld JJ,et al.Impact of urbanization on

detection rates of eating disorders.Am J Psychiatry .1995;152:

1272-1278.