5
JOURNAL OF WOMEN’S HEALTH Volume 12, Number 5, 2003 © Mary Ann Liebert, Inc. Young Adults Admitted for Asthma: Does Gender Influence Outcomes? SCOTT E. WOODS, M.D., M.P.H., M.Ed., 1 JONATHAN SORSCHER, M.D., 1 JOHN KING, M.D., 1 and KIM HASSELFELD 2 ABSTRACT Objective: To assess any outcome differences between young men and women who are ad- mitted for asthma. Methods: We conducted a retrospective cohort study based on hospitalizations. An inclu- sion criterion was admission for asthma between January 1, 1998 and July 1, 2001. Exclusion criteria included age .45, chronic obstructive pulmonary disease (COPD), and emphysema. Data were collected on 10 potential confounding variables. Four outcome variables were as- sessed, including length of stay, intensive care unit (ICU) length of stay, mortality, and res- piratory failure. Results: Patients admitted for asthma were significantly more likely to be female (374 fe- males vs. 106 males, p,0.05). There was no difference between the genders comparing month of admission. The women were significantly older, with more Medicaid insured, and more anxiety/depression (p,0.05). There was no difference between the genders for obesity, race, tobacco history, gastroesophageal reflux disease (GERD), hypertension, diabetes, and pneu- monia. There was no reported mortality. Using regression analysis, there was no difference between the genders for length of stay (odds ratio [OR] 5 1.06, 95% confidence interval [CI] 0.97–1.17) and respiratory failure (OR 5 1.58, 95% CI 0.53–4.76). Men stayed significantly longer in the ICU (OR 5 1.18, 95% CI 1.01–1.38). Conclusions: Patients admitted with asthma are significantly more likely to be female. Males stay significantly longer in the ICU. There is no difference between the genders for length of stay and respiratory failure. There was no reported mortality for either gender. 481 INTRODUCTION D ESPITE RECENT ADVANCES in the management of asthma, this disease continues to cause significant morbidity and mortality in both chil- dren and adults. Young children have a high asthma prevalence, with approximately half of all asthma cases diagnosed in the prepubertal decade. In childhood, a 2:1 male/female preva- lence of asthma exists, but this reduces to a 1:1 ratio by age 30. 1 In children under age 10, where male prev- alence of the disease is nearly 2:1, males are hospitalized at a greater rate than females. 2–9 This difference in asthma hospitalizations subse- quently tends to disappear, and the rate of hos- pitalizations becomes equal in early puberty. 2,3 Through adulthood, even while the prevalence of 1 Bethesda Family Medicine Residency Program, Cincinnati, Ohio. 2 E. Kenneth Hatton Research Center, Good Samaritan Hospital, Cincinnati, Ohio.

Young Adults Admitted for Asthma: Does Gender Influence Outcomes?

  • Upload
    kim

  • View
    217

  • Download
    2

Embed Size (px)

Citation preview

Page 1: Young Adults Admitted for Asthma: Does Gender Influence Outcomes?

JOURNAL OF WOMEN’S HEALTHVolume 12, Number 5, 2003© Mary Ann Liebert, Inc.

Young Adults Admitted for Asthma: Does Gender Influence Outcomes?

SCOTT E. WOODS, M.D., M.P.H., M.Ed.,1 JONATHAN SORSCHER, M.D.,1

JOHN KING, M.D.,1 and KIM HASSELFELD2

ABSTRACT

Objective: To assess any outcome differences between young men and women who are ad-mitted for asthma.

Methods: We conducted a retrospective cohort study based on hospitalizations. An inclu-sion criterion was admission for asthma between January 1, 1998 and July 1, 2001. Exclusioncriteria included age .45, chronic obstructive pulmonary disease (COPD), and emphysema.Data were collected on 10 potential confounding variables. Four outcome variables were as-sessed, including length of stay, intensive care unit (ICU) length of stay, mortality, and res-piratory failure.

Results: Patients admitted for asthma were significantly more likely to be female (374 fe-males vs. 106 males, p,0.05). There was no difference between the genders comparing monthof admission. The women were significantly older, with more Medicaid insured, and moreanxiety/depression (p,0.05). There was no difference between the genders for obesity, race,tobacco history, gastroesophageal reflux disease (GERD), hypertension, diabetes, and pneu-monia. There was no reported mortality. Using regression analysis, there was no differencebetween the genders for length of stay (odds ratio [OR] 5 1.06, 95% confidence interval [CI]0.97–1.17) and respiratory failure (OR 5 1.58, 95% CI 0.53–4.76). Men stayed significantlylonger in the ICU (OR 5 1.18, 95% CI 1.01–1.38).

Conclusions: Patients admitted with asthma are significantly more likely to be female. Malesstay significantly longer in the ICU. There is no difference between the genders for lengthof stay and respiratory failure. There was no reported mortality for either gender.

481

INTRODUCTION

DESPITE RECENT ADVANCES in the managementof asthma, this disease continues to cause

significant morbidity and mortality in both chil-dren and adults. Young children have a highasthma prevalence, with approximately half of all asthma cases diagnosed in the prepubertaldecade. In childhood, a 2:1 male/female preva-

lence of asthma exists, but this reduces to a 1:1ratio by age 30.1

In children under age 10, where male prev-alence of the disease is nearly 2:1, males are hospitalized at a greater rate than females.2–9

This difference in asthma hospitalizations subse-quently tends to disappear, and the rate of hos-pitalizations becomes equal in early puberty.2,3

Through adulthood, even while the prevalence of

1Bethesda Family Medicine Residency Program, Cincinnati, Ohio.2E. Kenneth Hatton Research Center, Good Samaritan Hospital, Cincinnati, Ohio.

Page 2: Young Adults Admitted for Asthma: Does Gender Influence Outcomes?

disease is the same for both genders, females arehospitalized at a greater rate than males.2,8–10

Skobeloff et al.3 found that prior to age 11 themale/female admission ratio was 2:1. It equalizedin the 11–20 year age group, but after age 30, fe-male admissions for asthma outnumbered thoseof males by 2.5:1. Internationally, similar resultshave been observed. Prescott et al.10 noted a 1.7relative risk (RR) of female over male admissionamong adults in the Copenhagen area. In Eng-land, Hyndman et al.8 reported a trend towardincreased female admissions with progressingage.

The most comprehensive study examining gen-der and morbidity and mortality outcomes duringhospitalization for asthma was conducted by Tra-wick et al.,11 who carried out a 10-year retro-spective study of 103 patients admitted withasthma. They investigated gender differencesduring admission of “high-risk” asthma patients.They defined “high-risk” as those patients ad-mitted two or more times for asthma or patientsadmitted with status asthmaticus. Of the 103 pa-tients, 72% were women. Women were 3.5 yearsolder than men, and the authors found a trendtoward a longer length of stay in women (p 50.0554). They found no significant difference inany other outcome, except that men had worsehypercapnea as determined by arterial bloodgases (ABG), suggesting a worse degree of air-way obstruction.11 However, the blood gas dataare difficult to evaluate, as only 53% of theirasthma cohort had an ABG test performed.

In our study, we examined gender-related dif-ferences in young adults admitted with asthma,and like Trawick et al., we investigated outcomesduring hospitalization. Our cohort included allseverities of asthma, not just high-risk patients.We controlled for multiple confounders, includ-ing insurance status, age, anxiety/depression,and smoking. We hypothesized that women ad-mitted for asthma would possess similar out-comes compared with men admitted for asthma.

MATERIALS AND METHODS

We conducted a retrospective cohort studybased on hospitalizations. Criteria for inclusionin the cohort included a hospital admission in aTriHealth Hospital with a primary diagnosis ofasthma by ICD-9 code between January 1, 1998,and July 1, 2001, and a minimum patient age of

18. There are three TriHealth hospitals, all locatedin suburban or urban Cincinnati, Ohio. Exclusioncriteria included chronic obstructive pulmonarydisease (COPD), emphysema, and age .45. Thisage limit was set to further reduce any chance ofincluding patients with a component of COPD.Data were collected on three demographic vari-ables, age, race, and insurance status, and sevencomorbidities, obesity, diabetes mellitus, anxi-ety/depression, gastroesophageal reflux disease(GERD), pneumonia, hypertension, and tobaccohistory. The outcomes of interest were mortality,total length of stay, intensive care unit (ICU)length of stay, and respiratory failure. Individu-als could join the asthma cohort only one timeduring the study period. All subsequent admis-sions were excluded. Readmissions were rare(n 5 7). Institutional Review Board approval wasobtained prior to data collection.

Univariate analysis using chi-square and t testswas performed comparing gender with each ofthe three demographic variables and with each ofthe seven comorbidities. To generate the adjustedrisks of each outcome, we performed logistic re-gression analysis for dichotomous variables andlinear regression for continuous variables whilecontrolling for the 10 potential confounding vari-ables. Analysis was performed using STATA(STATA Corporation, College Station, TX) andBMDP New System (Statistical Solutions, Saugus,MA) statistical software. Using a two-tailed alphaof 0.05 and a beta of 0.10, it was estimated thatapproximately 432 patients would be required forthis study to have 90% power to find a significantdifference between the genders for any outcomeif they differed by 15%.

RESULTS

Four hundred eighty patients were availablefor analysis. Patients admitted for asthma weresignificantly more likely to be female (374 femalesvs. 106 males, p 5 0.001). There was no differencebetween the genders comparing proportions ad-mitted monthly (p 5 0.22) (Fig. 1). The womenwere significantly older (36.0 vs. 32.6 years, p 50.001), more of them were Medicaid insured (17%vs. 6%, p 5 0.01), and more experienced anxi-ety/depression (20% vs. 5%, p 5 0.001). Therewas no difference between the genders for obe-sity, race, tobacco history, GERD, hypertension,diabetes, and pneumonia. There was no reported

WOODS ET AL.482

Page 3: Young Adults Admitted for Asthma: Does Gender Influence Outcomes?

mortality for either gender. Using regressionanalysis, there was no difference between thegenders for length of stay (odds ratio [OR] 5 1.06,95% confidence interval [CI] 0.97–1.17) and res-piratory failure (OR 5 1.58, 95% CI 0.53–4.76).Men stayed significantly longer in the ICU (OR 51.18, 95% CI 1.01–1.38, p 5 0.03).

DISCUSSION

This study provides further evidence of a fe-male predominance among adults admitted withasthma. We found women to be 3.7 times morelikely to be admitted at our institution. This issimilar to the results found by Trawick et al.,11

where women made up 72% of the study popu-lation. Similarly, Skobeloff et al.3 found thatamong adults hospitalized with asthma, there isa female/male ratio of 3:1 among 20–50-year-oldsand 2.5:1 for those .50.

Interpreting this gender-related difference ischallenging, as there are numerous factors influ-encing admission to the hospital. Theories as towhy adult women are admitted more than men

despite an equal prevalence of disease have in-cluded differences in response to equivalent de-grees of airway narrowing,12–14 differences in rel-ative size of the airways compared with lungsize,15 correct use of metered dose inhalers,10,12,15

physician bias regarding use of steroids in youngwomen,3,10 physician difference in the educationand treatment of female patients,3,10,15 access tocare,15 hormone status,3,15,16 and disproportion-ate assumption of the domestic exposures of childcare, dusting, cleaning, and cooking.15 Theoriesas to why adult men are likely to be admitted forasthma included employment obligations, disen-chantment with professional care, and perhapsmore effective use of medications. There arelikely multiple factors influencing whether or notto admit a patient with asthma, and more stud-ies are needed to look at factors influencing thisdecision.

Measures of morbidity have been shown to bedisproportionately higher in females. These in-clude lower quality of life scales,12,13 higher nonhospitalization medical usage,12,17 longerhospital stays,3 and a higher readmission rate for pediatric female asthmatics.2,18 In a Sas-

GENDER AND ASTHMA IN YOUNG ADULTS 483

FIG. 1. Number of men and women admitted for asthma by month (1998–2001), p 5 0.22.

Page 4: Young Adults Admitted for Asthma: Does Gender Influence Outcomes?

katchewan study, Senthilselvan2 found a higherinitial admission rate for boys aged 10–14 years,although the girls in this age group had a 60%higher readmission rate. Similar results werefound in children aged ,1–14 in New Zealand byMitchell et al.,18 with females having 23% higherreadmissions. At least one study found that thetrend toward female readmission carries intoadulthood. Heard et al.19 found a female pre-dominance in asthma readmissions in adultsvarying from 55% to 75% in two hospitals in Aus-tralia.

We found women admitted with asthma tohave significantly higher rates of anxiety/de-pression than men, and this variable has not beenlooked at in previous studies. Women were alsosignificantly older and had more Medicaid in-surance than men in our study, which is similarto the findings of Trawick et al.11 However, con-trolling for all three variables with regressionanalysis did not change the outcomes measured.

Our hypothesis was that men and women ad-mitted for asthma would possess similar out-comes. However, in our study, men had worseoutcomes than women during hospitalization.We found men to have a significantly longer ICUlength of stay than women, a longer total lengthof stay, and more respiratory failure, but theselast two outcomes did not reach statistical sign-ificance. This suggests that the men, although less frequently admitted, tended to have more severe disease. This is similar to the finding ofTrawick et al.11 that men consistently have higherPCO2 values on ABG analysis, and this disparitymarkedly increased when comparing ABGs frompatients from the medical floor, medical ICU, andintubated patients, respectively. When looking atthe other morbidity outcomes, however, Trawicket al. found a nonsignificant trend in the oppo-site direction. Women had a longer total lengthof stay, more admissions to the ICU, and higherrates of endotracheal intubation in their study.11

These differences between the two studies maybe partially explained by differing patient popu-lations. Trawick et al. examined high-risk asthmapatients, whereas our study included a largerpopulation of broad-spectrum asthma.

A closer look at the prevalence of our measuredoutcomes suggests that our study populationpossessed less severe asthma. First, we had no re-ported mortality during the 31/2 years of thestudy. Nationally, the annual mortality rate from

asthma is approximately 5000/year.1 Trawick etal.,11 with a smaller population with more severeasthma, had two reported mortalities during theirstudy period. In addition, respiratory failure was uncommon. Although males developed 58%more respiratory failure than women in ourstudy, it did not reach statistical significance be-cause of the low prevalence of the outcome (14 of374 women [3.7%], 6 of 106 men [5.6%]). A powercalculation on this outcome with a standard al-pha, beta and with 480 patients would yield 11%power to find this difference statistically signifi-cant. Approximately 7000 patients with the sameprevalence of disease and proportion in the two genders would be required to find a signifi-cant difference. Although readmissions were ex-cluded, they were rare in our study population(n 5 7), also suggesting that our study populationhad milder disease.

There are limitations to our study that must betaken into account when considering our results.This was a retrospective study, so the patientscould not be followed over time to determinelong-term outcomes. We relied on ICD-9 codesfor our diagnoses and in hospital outcomes, andit is possible that patients may have been codedinaccurately (e.g., underreporting of intubation/respiratory failure). Also, we did not request ac-cess to charts and, therefore, could not look at in-dicators of disease severity, such as ABG analy-ses, peak flow rates, types of chronic medicationsprescribed, and differences in pulmonary func-tion testing. These variables may have helped tobetter understand the study population and theextent of their asthma.

Future studies might look prospectively at menand women before hospitalization, during hospi-talization, and at follow-up to ascertain long-termoutcomes. A larger study, possibly using statewidedata, examining gender and morbidity and mor-tality outcomes for hospitalized adult patients withasthma could help clarify these different observa-tions. In addition, there have not been any pedi-atric studies examining gender and hospitalizationoutcomes for admitted patients with asthma.

In conclusion, young adults admitted withasthma were significantly more likely to be wo-men. Men stayed significantly longer in the ICU,had a longer total length of stay, and experiencedmore respiratory failure. However, these differ-ences did not reach statistical significance. Therewas no mortality for either gender.

WOODS ET AL.484

Page 5: Young Adults Admitted for Asthma: Does Gender Influence Outcomes?

REFERENCES

1. McFadden ER Jr. Asthma. In: Braunwald E, Fauci AS,Kasper DL, Hauser SL, Longo DL, Jameson JL, eds.Harrison’s principles of internal medicine, 15th ed.New York: McGraw-Hill, 2001:1456.

2. Senthilselvan A. Effect of readmissions on increasinghospital admissions for asthma in children. Thorax1995;50:934.

3. Skobeloff EM, Spivey WH, St. Clair SS, Schoffstall JM.The influence of age and sex on asthma admissions.JAMA 1992;268:3437.

4. Conway SP, Littlewood JM. Admission to hospitalwith asthma. Arch Dis Child 1985;60:636.

5. Gerstman BB, Bosco LA, Tomita DK. Trends in theprevalence of asthma hospitalization in the 5- to 14-year-old Michigan Medicaid population, 1980 to 1986.J Allergy Clin Immunol 1993;91:838.

6. Evans R, Mullally DI, Wilson RW, et al. National trendsin the morbidity and mortality of asthma in the U.S.:Prevalence, hospitalization and death from asthma overtwo decades: 1965–1984. Chest 1987;91(Suppl):65S.

7. Gergen PJ, Weiss KB. Changing patterns of asthmahospitalization among children: 1979 to 1987. JAMA1990;264:1688.

8. Hyndman SJ, Williams DRR, Merrill SL, LipscombeJM, Palmer CR. Rates of admission to hospital forasthma. Br Med J 1994;308:1596.

9. Wilkins K, Mao Y. Trends in rates of admission to hos-pital and death from asthma among children andyoung adults in Canada during the 1980s. Can MedAssoc J 1993;148:185.

10. Prescott E, Lange P, Vestbo J, Copenhagen City HeartStudy Group. Effect of gender on hospital admissionsfor asthma and prevalence of self-reported asthma: Aprospective study based on a sample of the generalpopulation. Thorax 1997;52:287.

11. Trawick DR, Holm C, Wirth J. Influence of gender onrates of hospitalization, hospital course, and hyper-capnea in high-risk patients admitted for asthma.Chest 2001;119:115.

12. Osborne ML, Vollmer WM, Linton KLP, Buist AS.Characteristics of patients with asthma within a largeHMO: A comparison by age and gender. Am J RespirCrit Care Med 1998;157:123.

13. Bousquet J, Knani J, Dhivert H, et al. Quality of lifein asthma: I. Internal consistency and validity of theSF-36 questionnaire. Am J Respir Crit Care Med 1994;149:371.

14. Turner MO, Noertjojo K, Vedal S, Bai T, Crump S,Fitzgerald JM. Risk factors for near-fatal asthma: A case-control study in hospitalized patients withasthma. Am J Respir Crit Care Med 1998;157:1804.

15. Redline S, Gold D. Challenges in interpreting genderdifferences in asthma. Am J Respir Crit Care Med1994;150:1219.

16. Rubio RL, Rodriguez GB, Collazo JJ. Comparativestudy of progesterone, estradiol, and cortisol concen-trations in asthmatic and nonasthmatic women. Al-lergol Immunopathol (Madr) 1988;16:263.

17. Butz AM, Eggleston P, Alexander C, Rosenstein BJ.Outcomes of emergency room treatment of childrenwith asthma. J Asthma 1991;28:255.

18. Mitchell EA, Bland JM, Thompson JMD. Risk factorsfor readmission to hospital for asthma in childhood.Thorax 1994:49:33.

19. Heard AR, Campbell DA, Ruffin RE, Smith B, LukeCG, Roder DM. Rehospitalisation for asthma within12 months: Unequal rates on the basis of gender attwo hospitals. Aust NZ J Med 1997;27:669.

Address reprint requests to:Scott E. Woods, M.D.

Director of EpidemiologyBethesda Family Medicine Residency Program

4411 Montgomery Road, Suite 200Cincinnati, OH 45212

E-mail: [email protected]

GENDER AND ASTHMA IN YOUNG ADULTS 485