THE EPIDEMIOLOGY OF APPENDICITIS AND APPENDECTOMY IN ...

16
AMERICAN JOURNAL OF EPIDEMIOLOGY VOL 132, No 5 Copyright © 1990 by The Johns Hopkins University School of Hygiene and Public Health Printed in U S A All rights reserved THE EPIDEMIOLOGY OF APPENDICITIS AND APPENDECTOMY IN THE UNITED STATES DAVID G. ADDISS, NATHAN SHAFFER, BARBARA S. FOWLER, AND ROBERT V. TAUXE Addlss, D. G. (CDC, Atlanta, GA 30333), N. Shaffer, B. S. Fowler, and R. V. Tauxe. The epidemiology of appendicitis and appendectomy in the United States. Am J Epidemiol 1990;132:910-25. To describe the epidemiology of appendicitis and appendectomy in the United States, the authors analyzed National Hospital Discharge Survey data for the years 1979-1984. Approximately 250,000 cases of appendicitis occurred annually in the United States during this period, accounting for an estimated 1 million hospital days per year. The highest incidence of primary positive appendectomy (appendicitis) was found in persons aged 10-19 years (23.3 per 10,000 population per year); males had higher rates of appendicitis than females for all age groups (overall rate ratio, 1.4:1). Racial, geographic, and seasonal differences were also noted. Appendicitis rates were 1.5 times higher for whites than for nonwhites, highest (15.4 per 10,000 population per year) in the west north central region, and 11.3% higher in the summer than in the winter months. The highest rate of incidental appendectomy was found in women aged 35-44 years (43.8 per 10,000 population per year), 12.1 times higher than the rate for men of the same age. Between 1970 and 1984, the incidence of appendicitis decreased by 14.6%; reasons for this decline are unknown. A life table model suggests that the lifetime risk of appendicitis is 8.6% for males and 6.7% for females; the lifetime risk of appendectomy Is 12.0% for males and 23.1% for females. Overall, an estimated 36 incidental procedures are performed to prevent one case of appendicitis; for the elderly, the preventive value of an incidental procedure is considerably lower. appendectomy; appendicitis; surgery Appendectomy for acute appendicitis is Much has been written on appendicitis one of the most frequently performed sur- since it was described by Fitz more than gical procedures in the United States (1). 100 years ago (2), but the etiology and epidemiology of this disease remain poorly understood. Acute appendicitis has been Received for publication November 21, 1989, and attributed to a variety of possible causes, infinalform April 16,1990. including mechanical obstruction (3, 4), in- Abbreviations: ICDA-8, Eighth Revision I n t e m a - , , ,., ,-, / r o \ r - i - i tional Classjication of Diseases, Adapted for Use m the adequate dietary fiber (5-8), familial SUS- United States; ICD-9-CM, International Classification ceptibility (9), factors associated with im- of Diseases, Ninth Revision, Clinical Modification. proved SOCioeconomic Conditions (10, 11), From the Enteric Diseases Branch, Division of ,, , . , . , , ... ,, Bacterial Diseases, Center for Infectious Diseases, and bacterial, Viral, and parasitic pathogens CentersforDisease Control, Atlanta, GA. (12-15). Few population-based epidemio- Reprint requests to Dr. David G. Addiss, Mailstop \ og [ c studies of appendicitis have been pub- C-09, Center for Infectious Diseases, Centers for Dis- ,. , , , , ,.,,, . , f ease Control, 1600 Clifton Road NE, Atlanta, GA ^^^ however, and little 18 known about 30333. the demographic characteristics, geo- The authors thank Dr. Nancy Hargrett-Bean for graphic differences, seasonality, Or long- statistical advice, Dr. Richard Dicker for assistance , , , , , . ... . ,, with the life table, and Bertha Smith for assistance term secular trends of appendicitis in the with manuscript preparation. United States. 910 at McGill University Libraries on October 10, 2013 http://aje.oxfordjournals.org/ Downloaded from at McGill University Libraries on October 10, 2013 http://aje.oxfordjournals.org/ Downloaded from at McGill University Libraries on October 10, 2013 http://aje.oxfordjournals.org/ Downloaded from at McGill University Libraries on October 10, 2013 http://aje.oxfordjournals.org/ Downloaded from at McGill University Libraries on October 10, 2013 http://aje.oxfordjournals.org/ Downloaded from at McGill University Libraries on October 10, 2013 http://aje.oxfordjournals.org/ Downloaded from at McGill University Libraries on October 10, 2013 http://aje.oxfordjournals.org/ Downloaded from at McGill University Libraries on October 10, 2013 http://aje.oxfordjournals.org/ Downloaded from at McGill University Libraries on October 10, 2013 http://aje.oxfordjournals.org/ Downloaded from at McGill University Libraries on October 10, 2013 http://aje.oxfordjournals.org/ Downloaded from at McGill University Libraries on October 10, 2013 http://aje.oxfordjournals.org/ Downloaded from at McGill University Libraries on October 10, 2013 http://aje.oxfordjournals.org/ Downloaded from at McGill University Libraries on October 10, 2013 http://aje.oxfordjournals.org/ Downloaded from at McGill University Libraries on October 10, 2013 http://aje.oxfordjournals.org/ Downloaded from at McGill University Libraries on October 10, 2013 http://aje.oxfordjournals.org/ Downloaded from at McGill University Libraries on October 10, 2013 http://aje.oxfordjournals.org/ Downloaded from

Transcript of THE EPIDEMIOLOGY OF APPENDICITIS AND APPENDECTOMY IN ...

Page 1: THE EPIDEMIOLOGY OF APPENDICITIS AND APPENDECTOMY IN ...

AMERICAN JOURNAL OF EPIDEMIOLOGY VOL 132, No 5

Copyright © 1990 by The Johns Hopkins University School of Hygiene and Public Health Printed in U S AAll rights reserved

THE EPIDEMIOLOGY OF APPENDICITIS AND APPENDECTOMY INTHE UNITED STATES

DAVID G. ADDISS, NATHAN SHAFFER, BARBARA S. FOWLER, ANDROBERT V. TAUXE

Addlss, D. G. (CDC, Atlanta, GA 30333), N. Shaffer, B. S. Fowler, and R. V.Tauxe. The epidemiology of appendicitis and appendectomy in the United States.Am J Epidemiol 1990;132:910-25.

To describe the epidemiology of appendicitis and appendectomy in the UnitedStates, the authors analyzed National Hospital Discharge Survey data for theyears 1979-1984. Approximately 250,000 cases of appendicitis occurred annuallyin the United States during this period, accounting for an estimated 1 millionhospital days per year. The highest incidence of primary positive appendectomy(appendicitis) was found in persons aged 10-19 years (23.3 per 10,000 populationper year); males had higher rates of appendicitis than females for all age groups(overall rate ratio, 1.4:1). Racial, geographic, and seasonal differences were alsonoted. Appendicitis rates were 1.5 times higher for whites than for nonwhites,highest (15.4 per 10,000 population per year) in the west north central region,and 11.3% higher in the summer than in the winter months. The highest rate ofincidental appendectomy was found in women aged 35-44 years (43.8 per 10,000population per year), 12.1 times higher than the rate for men of the same age.Between 1970 and 1984, the incidence of appendicitis decreased by 14.6%;reasons for this decline are unknown. A life table model suggests that the lifetimerisk of appendicitis is 8.6% for males and 6.7% for females; the lifetime risk ofappendectomy Is 12.0% for males and 23.1% for females. Overall, an estimated36 incidental procedures are performed to prevent one case of appendicitis; forthe elderly, the preventive value of an incidental procedure is considerably lower.

appendectomy; appendicitis; surgery

Appendectomy for acute appendicitis is Much has been written on appendicitisone of the most frequently performed sur- since it was described by Fitz more thangical procedures in the United States (1). 100 years ago (2), but the etiology and

epidemiology of this disease remain poorlyunderstood. Acute appendicitis has been

Received for publication November 21, 1989, and attributed to a variety of possible causes,in final form April 16,1990. including mechanical obstruction (3, 4), in-

A b b r e v i a t i o n s : I C D A - 8 , E i g h t h R e v i s i o n I n t e m a - , , , . , , - , / r o \ r - i - itional Classjication of Diseases, Adapted for Use m the adequate dietary fiber (5-8), familial SUS-United States; ICD-9-CM, International Classification ceptibility (9), factors associated with im-of Diseases, Ninth Revision, Clinical Modification. proved SOCioeconomic Conditions (10, 11),

From the Enteric Diseases Branch, Division of , , , . , . , , . . . ,,Bacterial Diseases, Center for Infectious Diseases, and bacterial, Viral, and parasitic pathogensCenters for Disease Control, Atlanta, GA. (12-15). Few population-based epidemio-

Reprint requests to Dr. David G. Addiss, Mailstop \og[c studies of appendicitis have been pub-C-09, Center for Infectious Diseases, Centers for Dis- ,. , , , , , . , , , . , fease Control, 1600 Clifton Road NE, Atlanta, GA ^ ^ ^ however, and little 18 known about30333. the demographic characteristics, geo-

The authors thank Dr. Nancy Hargrett-Bean for graphic differences, seasonality, Or long-statistical advice, Dr. Richard Dicker for assistance , , , , ,. . . . . ,,with the life table, and Bertha Smith for assistance term s e c u l a r t r e n d s o f appendicitis in thewith manuscript preparation. United States.

910

at McG

ill University L

ibraries on October 10, 2013

http://aje.oxfordjournals.org/D

ownloaded from

at M

cGill U

niversity Libraries on O

ctober 10, 2013http://aje.oxfordjournals.org/

Dow

nloaded from

at McG

ill University L

ibraries on October 10, 2013

http://aje.oxfordjournals.org/D

ownloaded from

at M

cGill U

niversity Libraries on O

ctober 10, 2013http://aje.oxfordjournals.org/

Dow

nloaded from

at McG

ill University L

ibraries on October 10, 2013

http://aje.oxfordjournals.org/D

ownloaded from

at M

cGill U

niversity Libraries on O

ctober 10, 2013http://aje.oxfordjournals.org/

Dow

nloaded from

at McG

ill University L

ibraries on October 10, 2013

http://aje.oxfordjournals.org/D

ownloaded from

at M

cGill U

niversity Libraries on O

ctober 10, 2013http://aje.oxfordjournals.org/

Dow

nloaded from

at McG

ill University L

ibraries on October 10, 2013

http://aje.oxfordjournals.org/D

ownloaded from

at M

cGill U

niversity Libraries on O

ctober 10, 2013http://aje.oxfordjournals.org/

Dow

nloaded from

at McG

ill University L

ibraries on October 10, 2013

http://aje.oxfordjournals.org/D

ownloaded from

at M

cGill U

niversity Libraries on O

ctober 10, 2013http://aje.oxfordjournals.org/

Dow

nloaded from

at McG

ill University L

ibraries on October 10, 2013

http://aje.oxfordjournals.org/D

ownloaded from

at M

cGill U

niversity Libraries on O

ctober 10, 2013http://aje.oxfordjournals.org/

Dow

nloaded from

at McG

ill University L

ibraries on October 10, 2013

http://aje.oxfordjournals.org/D

ownloaded from

at M

cGill U

niversity Libraries on O

ctober 10, 2013http://aje.oxfordjournals.org/

Dow

nloaded from

Page 2: THE EPIDEMIOLOGY OF APPENDICITIS AND APPENDECTOMY IN ...

APPENDICITIS AND APPENDECTOMY IN THE UNITED STATES 911

Incidental appendectomies are com-monly performed at the time of other ab-dominal or pelvic surgery to prevent futureappendicitis (16). Because the epidemiol-ogy of incidental appendectomy for theUnited States as a whole has not been wellcharacterized, the impact of this surgicalpractice is difficult to assess. Morbidityfrom the procedure is generally consideredto be negligible (17); however, the preven-tive value of incidental appendectomy, par-ticularly in the elderly, has been the subjectof recent debate (16, 18).

We analyzed 15 years of data from theNational Hospital Discharge Survey to de-scribe the epidemiology of appendicitis andincidental appendectomy in the UnitedStates. Using a life table model, we estimatethe current lifetime risk of appendicitis inthe United States and describe the preven-tive value of incidental appendectomiesperformed in persons of different ages.

MATERIALS AND METHODS

Since 1963, the National Center forHealth Statistics has conducted the Na-tional Hospital Discharge Survey, whichprovides data on a representative 0.5 per-cent sample of patients hospitalized in non-federal acute-care facilities in the UnitedStates (1). Each year, approximately

200,000 discharge records are randomly se-lected for review from a sample of 289-432hospitals stratified by size (number of beds)and geographic region (19). Stratum-specific weights are applied to derive na-tional estimates. Information obtainedfrom the hospital records includes patientdemographic characteristics, diagnoses,surgical procedures performed, dischargestatus, dates of admission and discharge,hospital characteristics, and region of thecountry. Up to seven diagnoses and foursurgical procedures are coded for each pa-tient, using the Eighth Revision Interna-tional Classification of Diseases, Adapted forUse in the United States (ICDA-8) (20) forthe years 1970-1978 and the InternationalClassification of Diseases, Ninth Revision,Clinical Modification (ICD-9-CM) (21)after 1978. The degree to which dischargediagnoses are confirmed by pathologic orlaboratory findings is unknown.

Data tapes were obtained for the years1970-1984. Primary appendectomy was de-fined as a nonincidental appendectomyprocedure code, either 41.1 (ICDA-8) or47.0 (ICD-9-CM) (table 1). A primary ap-pendectomy was defined as positive if thepatient also had a discharge diagnosis ofacute appendicitis with peritonitis (ICDA-8 diagnostic code 540.0 and ICD-9-CM di-

TABLE l

Diagnostic and surgical procedure codes for appendicitis and appendectomy, from the Eighth RevisionInternational Classification of Diseases, Adapted for Use in the United States (ICDAS) and the International

Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM)

ICDA-8 (1970-1978) ICD-9-CM (1979-1984)

Diagnostic codes

540 Acute appendicitis640.0 With peritonitis (abscess, perforation,

peritonitis, or rupture)

540.9 Without mention of peritonitis541 Appendicitis, unqualified

540 Acute appendicitis640.0 With generalized peritonitis (per-

foration, peritonitis, or rupture)640.1 With peritoneal abscess540.9 Without mention of peritonitis541 Appendicitis, unqualified

Procedure codes

41 Operations on appendix41.1 Appendectomy

47 Operations on appendix47.0 Appendectomy, excludes incidental47.1 Incidental appendectomy

Page 3: THE EPIDEMIOLOGY OF APPENDICITIS AND APPENDECTOMY IN ...

912 ADDISS ET AL.

agnostic codes 540.0 and 540.1), acute ap-pendicitis without mention of peritonitis(code 540.9), or appendicitis, unqualified(code 541). A patient with a positive pri-mary appendectomy was therefore consid-ered to have acute appendicitis; the termsare used interchangeably in this paper. Aprimary appendectomy was defined as neg-ative if none of these diagnostic codes foracute appendicitis was recorded. Negativeappendectomies, which included the diag-noses of chronic or recurrent appendi-citis and appendiceal lymphoid hyper-plasia, were considered to represent proce-dures performed on patients who had beensuspected of having appendicitis (since in-cidental appendectomy was not specified)but were found at the time of surgery notto have acute appendicitis.

An incidental appendectomy was definedas procedure code 47.1 in the ICD-9-CM(table 1); incidental appendectomies werenot recorded before 1979. Diagnostic accu-racy was defined as the proportion of allprimary appendectomies that were positive,and was considered equivalent to the posi-tive predictive value of the surgeon's pre-operative diagnosis. The perforation ratiowas the percentage of primary positive ap-pendectomies with evidence of perforation,peritonitis, rupture, or abscess (ICDA-8 di-agnostic code 540.0 and ICD-9-CM diag-nostic codes 540.0 and 540.1). The case-fatality ratio was defined as the percentageof patients with appendectomy who diedduring hospitalization.

In evaluating temporal trends, we usedUS resident population estimates for theyears 1970-1984 to calculate annual ratesof appendicitis and appendectomy (22, 23).For all other analyses, the mean annualincidence for the years 1979-1984 was de-termined by combining discharges for theseyears and using the 1980 US Census datafor the denominator (24). Regional com-parisons were based on the nine geographicdivisions of the US Census (22-24).

To determine lifetime and 1-year risk ofappendicitis and appendectomy in persons

with an intact appendix, we constructed alife table in 5-year age intervals using com-bined incidence data from 1979 to 1984(25). In the life table analysis, persons whohad had an appendectomy were consideredno longer at risk of appendicitis and wereexcluded from the denominator of succes-sive age groups. Lifetime risk was calcu-lated assuming a life span of 75 years andan incidence of appendicitis and appendec-tomy that remained constant at 1979-1984levels.

SAS statistical software was used to ana-lyze the data (26). The Pearson correlationcoefficient (r) was used to determine thedegree of linear correlation between vari-ables. No generally acceptable methods ex-ist for calculating confidence intervals forrates and proportions over multiple yearsfor these data; values are therefore pre-sented as point estimates. However, foreach year from 1970 to 1984, the relativestandard error for the number of appendec-tomies and appendicitis cases varied from5 percent to 9 percent (19).

RESULTS

In the period 1979-1984, an estimated3.4 million appendectomies were performedin the United States (561,000 per year), acrude annual incidence of 26 per 10,000population. Of these, 53 percent were pri-mary appendectomies; 85.3 percent ofpatients with primary appendectomy(251,000 per year) had a discharge diagnosisof acute appendicitis (primary positive ap-pendectomy). The crude incidence of acuteappendicitis was 11 per 10,000 populationper year. Forty-seven percent of appendec-tomies were incidental; the annual rate ofincidental appendectomy was 12 per 10,000population. These national figures werebased on a total of 16,457 hospital recordswith a procedure code of appendectomy.

The case-fatality ratio for both primarypositive and primary negative appendec-tomy was 0.3 percent. The case-fatality ra-tio for primary positive appendectomy was4.6 percent in persons aged 65 years or more

Page 4: THE EPIDEMIOLOGY OF APPENDICITIS AND APPENDECTOMY IN ...

APPENDICITIS AND APPENDECTOMY IN THE UNITED STATES 913

and 0.2 percent in persons aged less than65 years.

Between 1979 and 1984, appendicitis ac-counted for an estimated 1 million hospitaldays annually. The median length of hos-pital stay was 4 days for patients with pri-mary appendectomy, 9 days if the appendixwas perforated, and 7 days for incidentalappendectomy performed at the time ofanother surgical procedure.

Primary positive appendectomy(appendicitis)

The age-specific incidence of acute ap-pendicitis followed a similar pattern formales and females, but males had higherrates at virtually all ages, with an overallmale:female rate ratio of 1.4:1 (figure 1).The incidence was highest in males aged10-14 years (27.6 per 10.000 population peryear) and in females aged 15-19 years (20.5per 10,000 population per year). In personsaged 45 years or more, appendicitis ratesremained relatively constant at approxi-mately six per 10,000 for males and fourper 10,000 for females. The median age forboth males and females with primary pos-itive appendectomy was 21 years; 69 per-cent of persons with appendicitis were lessthan 30 years old.

Primary negative appendectomy anddiagnostic accuracy

The incidence of primary negative ap-pendectomy was higher in females than inmales, and was highest among women inthe childbearing years (figure 1). The rateof negative appendectomy among femalesaged 15-24 years (4.9 per 10,000 populationper year) was 2.5 times higher than that formales of the same age. Overall diagnosticaccuracy was lower for females (78.6 per-cent) than for males (91.2 percent). In fe-males, diagnostic accuracy dropped sharplyduring the childbearing years; in males, itdid not vary appreciably with age (figure 2).

Diagnoses most commonly associatedwith primary negative appendectomy in-cluded mesenteric lymphadenitis, other dis-

eases of the appendix, and gynecologic con-ditions (table 2).

Incidental appendectomy

The incidence of incidental appendec-tomy was 6.6 times higher in females thanin males (figure 1). In females, 62.7 percentof all appendectomies were incidental, com-pared with 17.7 percent in males. Womenaged 35—44 years had the highest rate ofincidental appendectomy, 43.8 per 10,000population per year, and were 12.1 timesmore likely to have an incidental appendec-tomy than were men of the same age. Inmales, the annual incidence of incidentalappendectomy gradually increased withage, to a rate of 7.3 per 10,000 populationamong men aged 65 years or more. Medianage at incidental appendectomy was 34years for females and 47 years for males. Infemales, primary surgical procedures mostcommonly performed at the time of inci-dental appendectomy were abdominal hys-terectomy (45.0 percent), oophorectomy orsalpingectomy (37.5 percent), cholecystec-tomy (18.4 percent), excision of ovariantissue (7.2 percent), and cesarean section(4.9 percent). In males, the most commonprimary surgical procedures were cholecys-tectomy (36.6 percent), total or partial ex-cision of the intestine (11.8 percent), andinguinal hernia repair (4.9 percent).

Appendiceal perforation

A total of 19.2 percent of appendicitiscases (primary positive appendectomies) inmales and 17.8 percent in females involvedappendiceal perforation, rupture, abscess,or generalized peritonitis. The perforationratio was lowest among persons aged 20-24years (9.1 percent) and increased directlywith age to 51 percent in persons aged 65years or more (figure 3); children aged lessthan 5 years were also at increased risk. Incontrast, the age-specific incidence of ap-pendiceal perforation was highest amongpersons aged 10-14 years (3.5 per 10,000population per year); a gradual increase inincidence with age was observed for personsover age 35 (figure 3).

Page 5: THE EPIDEMIOLOGY OF APPENDICITIS AND APPENDECTOMY IN ...

914

M -|

ADDISS ET AL.

Positive Primary Appendectomies

io

0)oU

10-

0-4 5 1 10-14 15-18 10-14 15-14 13-44 4S-S4 S5-M 15-74 2 75

Negative Primary Appendectomies

« -5 -

g H^ i -

om5

i-

00-4 5-0 10-14 1MB 10-14 15-14 15-44 45-54 55-14 65-74 2 75

Incidental Appendectomies

0-4 5-1 10-14 15-11 10-24 15-14 15-44 45-54 55-14 15-74 2 75

Age Group

FIGURE 1. Annual incidence of appendectomy (per 10,000 population) in the United States, 1979-1984, byage group, sex, and type of appendectomy.

Race, region, and season appendectomy. The incidences of primarypositive, primary negative, and incidental

Race (white or nonwhite) was recorded appendectomy were each 1.4-1.6 timesfor 91 percent of all hospital discharges for higher for whites than for nonwhites, a

Page 6: THE EPIDEMIOLOGY OF APPENDICITIS AND APPENDECTOMY IN ...

APPENDICITIS AND APPENDECTOMY IN THE UNITED STATES 915

moa.

a>

aa>o.

0-4 10-14 15-19 20-24 25-14 15-44 45-54 55-64 65-74 S75

Age GroupFIGURE 2 Diagnostic accuracy of primary appendectomies performed in the United States, by age group

and sex, 1979-1984. Diagnostic accuracy was defined as the proportion of all primary appendectomies with adischarge diagnosis of appendicitis.

TABLE 2

Discharge diagnoses and surgical procedures most commonly recorded for persons with primary negativeappendectomy, National Hospital Discharge Survey, 1979-1984*

ICD-9-CM code

Diagnostic code543.9

543.0

542289.2789.0558.9

574620.2218617616.0

Procedure code65.3-65.665.2

68.3-68.451.254.554.154.2

Diagnosis or procedure

Other diseases of the appen-dixt

Lymphoid hyperplasia of theappendix

Other appendicitis!Mesenteric lymphadenitisAbdominal painGastroenteritis, noninfec-

tiousCholelithiasisOvarian cystUterine leiomyomaEndometriosisCervicitis, endocervicitis

OophorectomyExcision or destruction of

ovaryAbdominal hysterectomyCholecystectomyLysis of adhesionsLaparotomyDiagnostic laparotomy

%ofdischarges

23.1

2.518.219.410.6

4.85.18.87.56.65.0

14.6

5.314.4

6.95.01.72.8

Females

Nationalestimate

(unweighted no.)

41,169 (194)

4,485 (25)32,514 (150)34,629 (172)18,955 (99)

8,503 (44)9,161 (42)

15,684 (73)13,311 (56)11,811 (52)8,839 (35)

26,040 (119)

9,420 (43)25,664 (114)12,359 (54)8,879 (43)2,990 (16)4,995 (23)

% ofdischarge*

16.6

3.315.829.015.4

10.83.90000

0

004.52.32.90.4

Males

Nationalestimate

(unweighted no.)

13,645 (61)

2,693 (16)12,988 (67)23,825 (117)12,661 (61)

8,840 (40)3,182 (12)

0 (0)0 (0)0 (0)0 (0)

0 (0)

0 (0)0 (0)

3,728 (15)1,918 (9)2,395 (10)

309 (2)

* International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM).t Colic, concretion, diverticulum, fecalith, fistula, intussusception, mucocele, or stercolith.% Chronic, recurrent, relapsing, or subacute.

Page 7: THE EPIDEMIOLOGY OF APPENDICITIS AND APPENDECTOMY IN ...

916 ADDISS ET AL

difference that persisted when data werestratified by sex, age group, and region. Theperforation ratio for nonwhites was 21.8percent, compared with 18.2 percent forwhites. Diagnostic accuracy was 86.1 per-cent in nonwhites and 85.3 percent inwhites.

Regional differences, which persistedwhen data were stratified by racial group,were also observed. The annual incidenceof appendicitis was highest (15.4 per 10,000population) in the west north central statesand lowest (9.4 per 10,000 population) in

the middle Atlantic states (table 3). Diag-nostic accuracy, which was lowest (81.1 per-cent) in the west north central states andhighest (89.4 percent) in the middle Atlan-tic states, tended to be inversely related tothe incidence of primary appendectomy(r = 0.52) and positively correlated withthe perforation ratio (r = 0.37), but thesetrends were not statistically significant. Re-gions with the highest rates of primaryappendectomy were also highest for inci-dental appendectomy.

The incidence of primary appendectomy

0-4 10-14 15-18 20-24 25-14 15-44 45-54 55-M 15-74 273

Age Group

FIGURE 3. Appendicitis perforation rate and perforation ratio in the United States, by age group, 1979-1984.Perforation rate was defined as the mean annual incidence of perforated appendicitis per 10,000 population.Perforation ratio was defined as the percentage of primary positive appendectomies with appendiceal perfora-tion.

TABLE 3

Diagnostic accuracy, perforation ratio, and incidence of primary appendectomy, acute appendicitis, andincidental appendectomy (per 10,000 population per year), by US Census region, National Hospital Discharge

Survey, 1979-1984*

US Census regionPnmaiy

appendectomyrate

Appendicitisrate

Diagnosticaccuracy

Perforationratio

Incidentalappendectomy

rate

New EnglandMiddle AtlanticEast North CentralWest North CentralSouth AtlanticEast South AtlanticWest South CentralMountainPacific

Entire United States

14.210.611.419.111.614.114.915.513.2

13.0

12.59.49.7

15.49.5

11.812.413.411.6

11.1

87.889.485.781.182.283.683.586.688.1

85.3

18.219.318.417.718.816.318.816.320.6

18.6

11.25.9

12.319.111.716.918.210.67.3

11.8

* The numbers of hospital records sampled during this period with a diagnostic or procedure code for primaryappendectomy, acute appendicitis, or incidental appendectomy were 8,717, 7,463, and 7,740, respectively.

Page 8: THE EPIDEMIOLOGY OF APPENDICITIS AND APPENDECTOMY IN ...

APPENDICITIS AND APPENDECTOMY IN THE UNITED STATES 917

and appendicitis, but not incidental appen-dectomy, appeared to increase during thesummer months (figure 4). During themonths of May through August, the num-ber of cases of appendicitis (adjusted to31-day months) was 11.3 percent higherthan during the winter months of Novem-ber through February.

Secular trends, 1970-1984

Between 1970 and 1984, the overall in-cidence of primary appendectomy de-creased by 22.1 percent; declines were ob-served in positive (14.6 percent) andnegative (52.5 percent) appendectomy ratesfor both sexes (figure 5). The greatestchanges occurred in the younger, higher-risk groups; among persons aged 10-24years, the primary appendectomy rate de-creased 25.5 percent, from 32.6 cases per10,000 population per year to 24.3 cases per10,000 population per year. Among non-white racial groups, the incidence of pri-mary appendectomy declined slightly, from9.7 per 10,000 in 1970 to 9.2 per 10,000 in1984.

Diagnostic accuracy appeared to increasesteadily during these 15 years, from 74 per-cent to 83 percent in females and from 86percent to 92 percent in males. The perfo-ration ratio also increased, from 15.6 per-cent in 1970-1972 to 19.5 percent in 1982-1984. A stronger association between the

perforation ratio and diagnostic accuracywas noted for males (r = 0.50, p = 0.07)than for females (r = 0.31, p = 0.29); bothdiagnostic accuracy and perforation ratiotended to be higher among males (figure 6).

Between 1979 (when incidental appen-dectomies were first coded) and 1984, inci-dental appendectomy rates decreased by27.7 percent in females and 13.6 percent inmales (figure 5). In women, this decline wasgreatest (33.6 percent) between the ages of25 and 54 years.

Life table analysis: risk of appendicitis

Assuming a constant incidence of appen-dicitis and appendectomy at 1979-1984 lev-els, the lifetime risk for a child aged lessthan 5 years of having his or her appendixsurgically removed (primary or incidentalappendectomy) was 12.0 percent for malesand 23.1 percent for females (table 4). Thelifetime risk of appendicitis was 8.6 percentfor males and 6.7 percent for females; 2.9percent of males and 16.0 percent of fe-males could expect to undergo incidentalappendectomy.

The risk of appendectomy during thenext year of life for those who had not yethad their appendix removed was higher forfemales than for males (table 5). The high-est risk of appendectomy from any causewas found in females aged 35-39 years; sixof every 1,000 women in this age group

CO• ocato3OU

)

COCOCO

3oa>

aE

170-

150-

150-

140 -

1S0-

120-

Z 110

All Primary

Jan Fob Mar Apr Uay Jun Jul Aug 8»p Ool Nov D«o

Month

FIGURE 4. Numbers of pnmary appendectomies (in thousands) in the United States, by month, 1979-1984.Adjusted to 31-day monthly totals; annual data combined.

Page 9: THE EPIDEMIOLOGY OF APPENDICITIS AND APPENDECTOMY IN ...

918 ADDISS ET AL

eoa3O.O

aooo

V)a>MaO

2>-

24-

20-

1S-

I -

s-

1 -

—or - o —

Female Incidental

Male positrve

-o—O---O---OFemale positive

Mala Incidental

70 71 72 73 74 75 70 77 78 79 80 81 82 83 84

Year

FIGURE 6. Appendicitis and appendectomy trends (per 10,000 population) in the United States, 1970-1984,by year, sex, and type of appendectomy (primary positive, primary negative, and incidental).

25 -1

O 20

2oa. 15-

1070 90 100

Diagnostic Accuracy

FIGURE 6. Appendicitis perforation ratio and diagnostic accuracy in the United States, by year and sex,1970-1984. Standard deviations (error bare) and regression lines are shown for males (A) and females (O).Each point represents data (by sex) for a single year.

Page 10: THE EPIDEMIOLOGY OF APPENDICITIS AND APPENDECTOMY IN ...

APPENDICITIS AND APPENDECTOMY IN THE UNITED STATES 919

would be expected to undergo appendec-tomy during the next year, and five of theseprocedures would be incidental appendec-tomies. In contrast, the highest risk of ap-pendectomy in males was in the 10-14-year

age group; three of 1,000 males in this agegroup would be expected to have an appen-dectomy during the following year, and fewof these (4 percent) would be incidentalappendectomies.

TABLE 4

Cumulative lifetime risk (to age 75 years) for appendectomy per 10,000 population. National Hospital DischargeSurvey, 1979-1984

Agegroup(yean)

0-45-9

10-1415-1920-2425-2930-3436-3940-4445-1950-5455-5960-6465-6970-74

All appendectomies

Males

1,1981,1761,111

96883072863756348241635530023216194

Females

2,3102,2952,2452,1441,9891,8031,5671,3111,028

75951736926917586

Primaryappendectomies

Males

938927863722585489408344281231187155115

7744

Females

843832776673547454379310245198156120

895728

Primary positiveappendectomies

Males

861851792662536447375318259213174144106

7243

Females

666657606518418345290238195158127100

754822

Incidentalappendectomies

Males

2872742712662602512382272061891721471188550

Females

1,6021,5951,5951,5771,5261,4131,2351,033

80257236725118111958

TABLE 5

Risk of appendectomy in the next year of life (per 100,000 population) for persons who had not yet had theirappendix removed, fry 5-year age group, National Hospital Discharge Survey, 1979-1984

group(years)

0-45-9

10-1416-1920-2425-2930-3435-3940-4445-4950-5455-5960-6465-6970-7476-7980-84£85

Allappendectomies

Males

51146319304221196156173136127116139144136190192125101

Females

39130258389458567596640689516309206192181172178140

99

Primaryappendectomies

Males

24139307292204169133131102916581786589896743

Females

2412322226719516814213396877163655856726841

Primary positiveappendectomies

Males

22129279269186160120120

95806176695986835240

Females

201081872091521151068974645451545344656541

Incidentalappendectomies

Males

277

111217272342353650596670

101104

5968

Females

157

36122264409455608493429238143127123116106

7257

Page 11: THE EPIDEMIOLOGY OF APPENDICITIS AND APPENDECTOMY IN ...

920 ADDISS ET AL

The preventive value of each incidentalappendectomy performed in different agegroups can be estimated using the life table.For example, 1,000 incidental appendecto-mies could be expected to prevent 52 casesof appendicitis when performed in womenaged 15-19 years, 24 cases when performedin women aged 35-39 years, and eight caseswhen performed in women aged 60-64years. Between 1979 and 1984, approxi-mately 260,000 incidental appendectomieswere performed annually in persons under75 years of age. Using the life table model,which is age-adjusted, we can see that these260,000 procedures prevented an estimated7,300 future lifetime cases of acute appen-dicitis; 36 incidental appendectomies weretherefore performed for each case of appen-dicitis averted.

DISCUSSION

To describe the epidemiology of appen-dicitis and appendectomy in the UnitedStates, we analyzed longitudinal, nationallyrepresentative, population-based data fromthe National Hospital Discharge Survey.Previous studies of appendicitis in thiscountry have been based on hospital caseseries, for which accurate denominator datawere unavailable (27, 28), or on regional orstatewide data (16, 29-32).

Our data have several important limita-tions. The actual number of hospital dis-charges sampled in any given year was rel-atively small, and it was often not possibleto stratify data by variables of interest. Inaddition, the discharge records did notspecify whether acute appendicitis was asurgical or a pathologic diagnosis. To theextent that the diagnosis was not con-firmed, the incidence of acute appendicitismay have been overestimated and the neg-ative appendectomy rate underestimated,since surgical diagnoses are more likely tobe positive than those which are patholog-ically confirmed (33, 34).

It is also possible that some incidentalappendectomies were miscoded as nonin-cidental. For example, 24 percent of pri-

mary negative (nonincidental) appendec-tomies were performed at the time ofanother major procedure, primarily oopho-rectomy, abdominal hysterectomy, or cho-lecystectomy. In these cases, which was the"incidental" procedure? We accepted non-incidental appendectomy codes at facevalue, assuming that acute appendicitismay have been suspected preoperatively("rule out appendicitis") and that other pa-thology had been discovered at the time ofsurgery. In doing so, we may have overes-timated the number of primary negativeappendectomies (30) and underestimateddiagnostic accuracy. Preoperative diag-noses, which could have helped resolve thisissue, were not available. The diagnosticaccuracy of 85 percent found in this study,however, is well within the range reportedin hospital-based studies where pathologicfindings and preoperative diagnoses wereavailable (27).

Our findings using national data gener-ally confirm the observations of others inregional studies and hospital case series.The overall incidence of appendicitis in theUnited States, 1.1 per 1,000 population peryear, is comparable to the rates of 0.96-1.2per 1,000 reported from Connecticut, SouthCarolina, and California (16, 30, 31). Al-though the incidence appears to be declin-ing, appendicitis-related morbidity remainshigh, and deaths from appendicitis still oc-cur, particularly in the elderly (35, 36).

One of the most striking epidemiologicfeatures of appendicitis is the marked var-iation in incidence by age and sex. In bothmales and females, the highest rates wereobserved in persons aged 10-19 years, afinding consistent with those of severalother studies (16, 30, 31, 37-39). In males,however, the peak incidence appeared atages 10-14 years, compared with ages 15-19 years in females. The incidence of ap-pendicitis was consistently higher in malesthan in females, even when persons withprevious appendectomy were excluded fromthe denominator. Appendicitis rates 1.2- to2.3-fold higher for males have also beenreported by other investigators (30, 31, 38-

Page 12: THE EPIDEMIOLOGY OF APPENDICITIS AND APPENDECTOMY IN ...

APPENDICITIS AND APPENDECTOMY IN THE UNITED STATES 921

40), but this difference remains unex-plained. The elevated rates in males acrossall age groups suggest that hormonalchanges in females may not play as signif-icant an etiologic role as previously hypoth-esized (41).

In contrast, primary negative appendec-tomy was more common in females, withthe highest rates occurring in women ofchildbearing age. Accurate diagnosis is par-ticularly difficult in this group, since gy-necologic disease often mimics the symp-toms of acute appendicitis (42). In addition,early surgical intervention, which has re-cently been encouraged because of an in-creased risk of tubal infertility followingappendiceal perforation (43), may explainthe higher rates of primary negative appen-dectomy in females.

A slight but consistent increase in appen-dicitis was noted during the summermonths, a pattern found in some previousstudies (44-46) but not others (27, 37, 38).The reasons for this seasonal pattern areunknown; speculation has focused onchanges in atmospheric pressure (47) andon both increases (38) and decreases (45)in relative humidity. The increase in inci-dence during the summer months may alsoreflect an infectious etiology for acute ap-pendicitis, a hypothesis supported by simi-lar summer peaks of other enteric infec-tions (46) and by the association of an acuteappendiceal syndrome with Yersinia andother enteric pathogens (12-14).

The overall incidence of appendicitis ap-pears to have decreased by 15 percent since1970; the greatest declines were observedin the populations at highest risk. Decliningrates of appendicitis have been reported inthe United States and Europe since WorldWar II (48-52), but the reasons for this areunclear. Changes in nutrition and diet (51),increased use of antibiotics (49), improve-ments in socioeconomic status (50), andchanges in patterns of infectious diseaseand immunity (52) have all been proposedas possible explanations, but no causal as-sociations have been demonstrated. It isalso possible that the decrease in the rate

of primary positive appendectomy between1970 and 1984 reflects changing medical orsurgical practices rather than an actual de-cline in the rate of appendicitis. In contrastto this pattern in the United States andEurope, the incidence of acute appendicitisappears to be increasing in developingcountries where it has historically been low(5, 38, 53-55). The increase has been at-tributed to dietary changes (5, 55), im-provements in socioeconomic status andhygienic standards (11, 54), and better ac-cess to health services (56). Some of thesesame factors may help explain the relativestability in appendicitis rates among non-whites in the United States between 1970and 1984 (a decrease of 4 percent), com-pared with a 16 percent decline amongwhites.

Regional variation in the incidence ofappendicitis, independent of race or agedistribution, was also noted. Geographicvariation in appendicitis and appendec-tomy has been well documented and hasbeen associated with differences in latitude(57), medical and surgical practices (56, 58,59), appendicitis case definitions (56), andactual risk of disease (60). Whether ethnicor cultural differences contribute to thegeographic variation in the United Statesis unknown, although this possibility is sug-gested by anecdotal observations. For ex-ample, immigrants from Germany, whereappendectomy and appendicitis mortalityrates are exceptionally high (59), settledprimarily in the upper Midwest (61), theregion with the highest incidence of acuteappendicitis in the United States. Differ-ences in surgical practices may also haveplayed a role, since the regions with thehighest rates of appendicitis were generallyalso highest for negative and incidental ap-pendectomy. The precise relation betweensurgical practice and the incidence of ap-pendicitis is difficult to determine, how-ever, since both the diagnosis and the cureof acute appendicitis depend on surgery. Ifsome cases of appendicitis do in fact resolvewithout surgical intervention, as some havesuggested (32, 62), then surgical practice

Page 13: THE EPIDEMIOLOGY OF APPENDICITIS AND APPENDECTOMY IN ...

922 ADDISS ET AL.

could indeed contribute to the observedregional variation.

The classic signs and symptoms of ap-pendicitis do not reliably distinguish be-tween patients who will have positive ap-pendectomies and those who will havenegative appendectomies (27). Althoughimproved diagnostic accuracy has been re-ported with use of computer algorithms (63,64) and high-resolution ultrasound (65,66),clinical skill and experience remain the ba-sis for managing patients with suspectedappendicitis (67). The challenge to the sur-geon is to prevent appendiceal perforationby early operation in cases of true appen-dicitis, but at the same time make the di-agnosis with sufficient specificity to avoidunnecessary negative appendectomies. Be-cause the initial signs and symptoms ofappendicitis are not pathognomonic, theprevailing view has been that an acceptablylow perforation ratio can be achieved onlyby surgeons who operate early and there-fore perform a certain number of negativeappendectomies (68, 69); the optimal neg-ative appendectomy rate has been claimedto be as high as 23 percent (27). Othershave argued that a low perforation ratioalone is a poor indicator of the quality ofsurgical practice (70), especially if it isachieved at the expense of excessively highnumbers of negative appendectomies,which can be associated with significantmorbidity (71). Indeed, several studies haveshown that clinical training (72), intensivein-hospital observation (73), and use of en-hanced algorithms for the differential di-agnosis of abdominal pain (64) can improvediagnostic accuracy without a concomitantincrease in the perforation ratio.

The national data also suggested a cor-relation between diagnostic accuracy andthe perforation ratio when analyzed by re-gion or year; small numbers limited statis-tical power, however, particularly in theanalysis by region. Diagnostic accuracy andthe perforation ratio were both higheramong men than among women, a findingreported previously (74). Diagnostic accu-

racy and the total primary appendectomyrate appeared to be inversely related whendata were analyzed by region, suggestingthat in areas where the clinical "threshold"for operating is lower, diagnostic accuracyis reduced.

Although the incidence of appendicealperforation did not vary as markedly withage as did the incidence of appendicitis, thetwo curves tended to be parallel; the peakincidence for both perforation and appen-dicitis occurred in persons aged 10-14years. In contrast, the perforation ratio wasstrongly age-related, being highest in theelderly and the very young. This " J-shaped"pattern has been noted by other investiga-tors (70, 75), and is thought to reflect boththe increased diagnostic difficulty and theless timely surgical intervention for personsin these extreme age groups (75-77).

In the life table analysis, we assumed aconstant incidence of appendectomy andappendicitis at 1979-1984 levels. Becauseincidence appears to be declining, the cur-rent risk of appendectomy and appendicitisin the United States may have been over-estimated. Nonetheless, at rates observedfor 1979-1984, one in eight males and onein four females can expect to have theirappendix surgically removed during theirlifetime; approximately one half of theseprocedures will be incidental appendecto-mies. The appropriateness and preventivevalue of incidental appendectomy, particu-larly in the elderly, has recently been ques-tioned (16, 18, 78). Although it is generallyconsidered a benign procedure (17), mostincidental appendectomies are performedin persons over age 35, which is well pastthe age of greatest risk for appendicitis, andin females, who are at lower risk thanmales. Furthermore, it is possible that theadditional surgical and hospital costs ofincidental appendectomy may be more sub-stantial than is generally assumed (16, 79).

The question of appropriate surgical pol-icy is not whether incidental appendectomyprevents future appendicitis but whetherthe procedure should be performed in per-

Page 14: THE EPIDEMIOLOGY OF APPENDICITIS AND APPENDECTOMY IN ...

APPENDICITIS AND APPENDECTOMY IN THE UNITED STATES 923

sons at low risk of appendicitis. At leasttwo different ages, 35 years and approxi-mately 60 years, have been proposed as agesat which to restrict incidental appendec-tomy (18, 78). The life table model suggeststhat limiting incidental appendectomies topersons under 35 years of age would reducethe total number of incidental proceduresin the United States each year by 50 per-cent (130,000 operations), but it might re-sult in as many as 2,200 additional lifetimecases of appendicitis, including 880 appen-diceal perforations. In comparison, limitingincidental appendectomies to persons un-der 60 years of age would reduce the num-ber of procedures by only 8 percent andresult each year in an additional 130 life-time cases of appendicitis (64 with perfo-ration). To prevent a single lifetime case ofacute appendicitis in persons aged 35 and60 years, 59 and 166 incidental proceduresare required, respectively. Although theoverall rate of incidental appendectomy de-clined sharply between 1979 and 1984, littlechange was noted among the elderly, forwhom the procedure has the lowest preven-tive value. Additional data on the economicand health consequences of incidental ap-pendectomy policy are needed to furtherrefine and guide surgical practice; mean-while, the indications for this operation inolder individuals remain controversial.

Although acute appendicitis is a commoncondition that has been recognized for morethan a century, its etiology is poorly under-stood. In this study, several previously rec-ognized epidemiologic patterns were con-firmed: Persons aged 10-19 years have thehighest rates of acute appendicitis; appen-dicitis is more common during the summer,and the incidence of the disease appears tobe higher in males than in females, higherin whites than in nonwhites, and higher inpersons living in certain regions of theUnited States, particularly the upper Mid-west. In addition, the rates of appendicitisand appendectomy appear to be decreasing.

Important diagnostic and policy ques-tions remain unanswered. To what extent

is it possible, using new technologies, toimprove diagnostic accuracy, particularlyin women of childbearing age who have thehighest rates of primary negative appen-dectomy? How can the perforation ratioand appendicitis-related mortality best bereduced in the elderly? Under what circum-stances should incidental appendectomy beconsidered inappropriate?

Progress in answering these and otherquestions has been hampered by a lack ofclarity and consensus in defining appendi-citis, diagnostic accuracy, negative appen-dectomy, and incidental appendectomy. Wehave proposed simple definitions based onICD codes that could be used in other stud-ies. Further understanding of the etiologyand epidemiology of appendicitis in theUnited States will require carefully de-signed prospective studies using pathologi-cally confirmed diagnoses and standardizedcase definitions.

REFERENCES

1. National Center for Health Statistics Utilizationof short-stay hospitals, United States, 1981 an-nual summary. Washington, DC: US GPO, 1983.(Vital and health statistics, Series 13, no. 72)(DHHS publication no. (PHS)83-1733).

2. Fitz RH. Perforating inflammation of the vermi-form appendix with special reference to its earlydiagnosis and treatment Trans Assoc Am Physi-cians 1886,1:107-44.

3. Jones BA, Demetriades D, Segal I, et al. Thepresence of appendiceal fecaliths in patients withand without appendicitis. Ann Surg 1985;202:80-2.

4. Pieper R, Kager L, Tidefeldt U. Obstruction ofappendix vermiformis causing acute appendicitis.Acta Chir Scand 1981,148:63-71.

5. Burkitt DP. The aetiology of appendicitis. Br JSurg 1971;68:695-9.

6. Barker DJP, Morris J, Nelson M. Vegetable con-sumption and acute appendicitis in 59 areas inEngland and Wales. Br Med J 1986;292:927-30.

7. Arnbjornsson E. Acute appendicitis and dietaryfiber. Arch Surg 1983;118:868-70.

8. Brender JD, Weiss NS, Koepsell TD, et al. Fiberintake and childhood appendicitis. Am J PublicHealth 1985;75:399-400.

9. Andersson N, Griffiths H, Murphy J. Is appendi-citis familial? Br Med J 1979;2:697-8.

10. Andersson N, Cockcroft A, Murphy JF, et al.Acute appendicitis and social class. Br J ClinPract 1981:35:272-3.

11. Barker DJP, Osmond C, Golding J, et al. Acute

Page 15: THE EPIDEMIOLOGY OF APPENDICITIS AND APPENDECTOMY IN ...

924 ADDISS ET AL.

apf)endicitis and bathrooms in three samples ofBritish children. Br Med J 1988;296:956-8.

12. Attwood SEA, Mealy K, Cafferkey MT, et al.Yereinia infection and acute abdominal pain. Lan-cet 1987;l:529-33.

13. Madden NP, Hart CA. Streptococcus miUen inappendicitis in children. J Pediatr Surg 1985;20:6-7.

14. Tobe T. Inapparent virus infection as a trigger ofappendicitis. Lancet 1965,1:1343-6.

15. Bredesen J, Lauritzen AF, Knstiansen VB, et al.Appendicitis and enterobiasis in children. ActaChir Scand 1988;154:585-7

16. Sugimoto T, Edwards D. Incidence and costs ofincidental appendectomy as a preventive measure.Am J Public Health 1987;77:471-5.

17. Voitk AJ, Lowry JB. Is incidental appendectomya safe practice? Can J Surg 1988;31:448-51.

18 NockertsSR,DetmerDE,FrybackDG Incidentalappendectomy in the elderly9 No. Surgery 1980;88:301-6.

19. National Center for Health Statistics. Trends inhospital utilization, United States, 1966-86.Hyattsville, MD. National Center for HealthStatistics, 1989. (Vital and health statistics, Series13, no. 101) (DHHS publication no. (PHS)89-1762).

20. National Center for Health Statistics. Eighth re-vision international classification of diseases,adapted for use in the United States. Vols 1 and2. Washington, DC: US GPO, 1968. (PHS publi-cation no. 1693).

21. Health Care Financing Administration. Interna-tional classification of diseases, ninth revision,clinical modification. Vols 1-3. Washington, DC:US GPO, 1980. (DHHS publication no. (PHS)80-1260).

22. US Bureau of the Census Current populationreports: population estimates and projections. Es-timates of the population of the United States byage, sex, and race. Washington, DC: US GPO,1982. (Series P-26, no. 917).

23. US Bureau of the Census. Current populationreports: population estimates and projections. Es-timates of the population of the United States byage, sex, and race. Washington, DC: US GPO,1986. (Series P-25, no. 985).

24. US Bureau of the Census. 1980 census of popula-tion, vol 1. Characteristics of the population: gen-eral population characteristics, United Statessummary. Washington, DC: US GPO, 1983.

25. Kleinbaum DG, Kupper LL, Morgenstem H.Epidemiologic research: principles and quantita-tive methods. New York, NY: Van Nostrand Rein-hold Company, 1982.

26. SAS Institute, Inc. SAS user's guide: basics, ver-sion 5 ed. Cary, NC: SAS Institute, Inc, 1985.

27. Berry J, Malt RA. Appendicitis near its centenary.Ann Surg 1984;200:567-75.

28. Kazarian KK, Roeder WJ, Mereheimer WL. De-creasing mortality and increasing morbidity fromacute appendicitis. Am J Surg 1970;119:681-5

29. Detmer DE, Nevers LE, Sikes ED. Regional re-sults of acute appendicitis care. JAMA 1981;246:1318-20.

30. Luckmann R. Incidence and case fatality rates for

acute appendicitis in California: a population-based study of the effects of age. Am J Epidemiol1989;129:905-18.

31. Pearson RJC. Acute appendicitis in the NewHaven Standard Metropolitan Area in 1958 and1959. Conn Med 1964;28:807-10.

32. Lembcke PA. Measuring the quality of medicalcare through vital statistics based on hospitalservice areas. I Comparative study of appendec-tomy rates. Am J Public Health 1962;42:276-86.

33. Bourke JB, Holliday A, Balfour TW, et al. Rela-tionship between surgeons' macroscopic diagnosisat emergency appendicectomy and subsequenthistopathology. (Abstract). Gut 1985;26:A565-6.

34. Jones MW, Paterson AG. The correlation betweengross appearance of the appendix at appendicec-tomy and histological examination. Ann R CollSurg Engl 1988;70:93-4.

35. Owens BJ, Hamit HF. Appendicitis in the elderlyAnn Surg 1978; 187:392-6.

36. Peltokallio P, Tykka H. Evolution of the agedistribution and mortality of acute appendicitisArch Surg 1981;116:153-6.

37. Ashley DJB. Observations on the epidemiology ofappendicitis. Gut 1967,8:533-8.

38. Freud E, Pilpel D, Mares AJ. Acute appendicitisin childhood in the Negev region: some epidemi-ological observations over an 11-year period(1973-1983) J Pediatr Gastroenterol Nutr 1988;7:680-4.

39 Pieper R, Kager L. The incidence of acute appen-dicitis and appendectomy. Acta Chir Scand 1982;148:45-9.

40. Soreide O. Appendicitis: a study of incidence,death rates, and consumption of hospital re-sources. Postgrad Med J 1984;60:341-5.

41. Arnbjornsson E. Varying frequency of acute ap-pendicitis in different phases of the menstrualcycle. Surg Gynecol Obstet 1982;155:709-ll.

42. Nakhgevany KB, Clarke LE. Acute appendicitisin women of childbearing age. Arch Surg 1986;121:1053-5.

43. Mueller BA, Daling JR, Moore DE, et al. Appen-dectomy and the risk of tubal infertility. N EnglJ Med 1986:315:1506-8.

44. Wolkomir A, Kornak P, Elsakr M, et al. Seasonalvariation of acute appendicitis: a 56-year study.South Med J 1987;80:958-60.

45. Brumer M. Appendicitis: seasonal incidence andpostoperative wound infection. Br J Surg 1970;5753-9.

46. Brink CF, Prinsloo H, Van der Poel JS. Eie sei-soenvoorkoms van akute appendicitis. (In Afri-kaans). S Afr Med J 1985;68:156-8.

47. Cihaf M, Kabelka M. Vhv atmosferickeho tlakuna vyskyt akutni apendicitidy u deti. (In Czech).Rozhl Chir 1980;59:622-6.

48. Raguveer-Saran MK, Keddie NC. The falling in-cidence of appendicitis. Br J Surg 1980,67:681.

49. Noer T. Decreasing incidence of acute appendici-tis. Acta Chir Scand 1975;141.431-2.

50. Palumbo LT. Appendicitis: Is it on the wane? AmJ Surg 1969;98:702-3.

51. Arnbjornsson E, Asp N-G, Westin SI. Decreasingincidence of acute appendicitis with special ref-

Page 16: THE EPIDEMIOLOGY OF APPENDICITIS AND APPENDECTOMY IN ...

APPENDICITIS AND APPENDECTOMY IN THE UNITED STATES 925

erence to the consumption of dietary fiber ActaChir Scand 1982; 148:461-4

52. Barker DJP. Acute appendicitis and dietary fibre:an alternative hypothesis. Br Med J 1985;290:1125-7.

53. Ofili 0. Implications of the rising incidence ofappendicitis in Africans. Cent Afr J Med 1987;33:243-6.

64. Osman AA. Epidemiological study of appendicitisin Khartoum. Int Surg 1974;59:218-21.

55. Moore SW, Robbs JV. Acute appendicitis in theZulu: an emerging disease9 (Letter). S Afr Med J1979;55:700.

56. Yoshida Y, Yoshida K. The high rate of appen-dectomy in Japan. Med Care 1976;14:950-7.

57. Barker DJP, Liggins A. Acute appendicitis in nineBritish towns. Br Med J 1981;283:1083-5.

58. Kisch AI, Moustafa AT, Gross AJ, et aL An epi-demiologic approach to the study of the incidenceof surgical procedures. Med Care 1969;7:471-80.

59. Lichtner S, Pflanz M. Appendectomy in theFederal Republic of Germany: epidemiology andmedical care patterns. Med Care 1971;9:311-30.

60. Walker ARP, Walker BF. Appendectomy in SouthAfrican inter-ethnic school pupils. Am J Gas-troenterol 1987;82:219-22.

61. Morin MM, Pickle LW, Mason TJ. Geographicpatterns of ethnic groups in the United States.Am J Public Health 1984;74:133-9.

62. Savrin RA, Clausen K, Martin EW, et al. Chronicand recurrent appendicitis. Am J Surg 1979,137:355-7.

63. de Dombal FT, Leaper DJ, Horrocks JC, et al.Human and computer-aided diagnosis of abdom-inal pain: further report with emphasis on perfor-mance of clinicians. Br Med J 1974;l:376-80.

64. Adams ID, Chan M, Clifford PC, et al. Computer-aided diagnosis of acute abdominal pain: a multi-centre study. Br Med J 1986;293:800-4.

65. Puylaert JBCM, Rutgers PH, Lalisang RI, et al.A prospective study of ultrasonography in thediagnosis of appendicitis. N Engl J Med 1987;317:666-9.

66. Puylaert JBCM, Vermeijden RJ, van der WerfSDJ, et al. Incidence and sonographic diagnosisof bacterial ileocaecitis masquerading as appen-dicitis. Lancet 1989;2:84-6.

67. Schwartz SI. Tempering the technological diag-nosis of appendicitis. N Engl J Med 1987;317:703-4.

68. Hobaon T, Rosenman LD. Acute appendicitis:When is it right to be wrong? Am J Surg 1964;108:306-12.

69. Malt RA. The perforated appendix. N Engl J Med1986;315:1546-7.

70. Koepsell TD, Inui TS, Farewell VT. Factors af-fecting perforation in acute appendicitis. Surg Gy-necol Obstet 1981;153:508-10.

71. Chang FC, Hogle HH, Welling DR. The fate ofthe negative appendix. Am J Surg 1973;126:752-4.

72. Nuetra RR. Appendicitis: decreasing normal re-movals without increasing perforations. Med Care1978;16:956-61.

73. White JJ, Santdlana M, Haller JA. Intensive ln-hospital observation- a safe way to decrease un-necessary appendectomy Am Surg 1975;41:793-8.

74. Slater PE, Ben-Dom R, Rosenbluth MS. The pnceof better diagnostic accuracy in suspected appen-dicitis in Jerusalem. Eur J Epidemiol 1987,3:181-6.

75. Scher KS, Coil JA. Appendicitis: factors that in-fluence the frequency of perforation. South MedJ 1980;73:1561-3.

76. Brender JD, Marcuse EK, Koepsell TD, et al.Childhood appendicitis: factors associated withperforation. Pediatrics 1985;76:301-6.

77. Gilbert SR, Emmens RW, Putnam TC Appendi-citis in children. Surg Gynecol Obstet 1985;161:261-5.

78 Ludbrook J, Spears GFS The risk of developingappendicitis. Br J Surg 1965,52:856-8.

79. Hewitt D, Milner J, Le Riche WH. Incidentalappendectomy: a statistical appraisal Can MedAssoc J 1969;100.1075-81.