HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:
Guest Access | Sign In via User Name/Password

This Article Abstract Full Text (PDF) Free Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Article Archive Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (32) Citing Articles via Google Scholar Google Scholar Articles by Mitchell, I. Articles by Hessel, P. A. Search for Related Content PubMed PubMed Citation Articles by Mitchell, I. Articles by Hessel, P. A.

Near-Fatal Asthma^*

A Population-Based Study of Risk Factors

^* From the Child Health Research Unit (Drs. Mitchell and Tough, and Ms. Semple) and the Respiratory Research Group (Dr. Green), University of Calgary, Calgary, AB; and the Department of Public Health Sciences (Dr. Hessel), University of Alberta, Edmonton, AB, Canada.

Correspondence to: I. Mitchell, MD, FCCP, 1820 Richmond Rd SW, Calgary, AB, Canada, T2T 5C7; e-mail: ian.mitchell{at}calgaryhealthregion.ca

	Abstract

TOP Abstract Introduction Materials and Methods Results Discussion References

 
Background: The study of near-fatal asthma (NFA) may provide a means to further our understanding of fatal asthma. Studies of NFA often are derived from a single ICU rather than from a defined population. We therefore aimed to identify factors distinguishing NFA patients (cases) from those persons treated in an emergency department (ED) [ED control subjects] and in the community (community control subjects [CCs]).

Methods: This was a population-based case-control study conducted over 20 months of 45 NFA patients (age range, 5 to 50 years), 197 ED control subjects treated in an ED, and 303 CCs, all of whom were residents of Alberta.

Results: The age distribution was similar between NFA patients and control subjects, with the majority being < 22 years of age (NFA patients, 68.9%; ED control subjects, 71.3%; CCs, 60.7%). Those patients with NFA were significantly more likely to have received a diagnosis before 5 years of age (66.6%), compared to ED control subjects (39.6%) and CCs (28.7%). The NFA group was significantly more likely to report moderate-to-severe disease and more frequent symptoms than the other groups. Therapy with bronchodilators was used most frequently by the NFA group compared to the ED control subjects and CCs (p < 0.001), as was therapy with inhaled steroids (p < 0.001) and oral steroids (p < 0.001). NFA patients had higher scores for vulnerability and were most likely to admit to stress as an asthma trigger. All groups had high exposure to cigarette smoke and pets.

Conclusion: NFA patients have many modifiable risk factors and many similarities to ED control subjects and CCs with asthma. General measures to improve asthma control and awareness of risks are required in all groups.

Key Words: asthma • near-fatal asthma • population based • risk factors

	Introduction

TOP Abstract Introduction Materials and Methods Results Discussion References

 
Asthma morbidity and mortality have risen since the 1970s.^{1 2} Studies of asthma deaths^{3 4 5 6 7 8} have suggested that a lack of recognition of disease severity by patients and undertreatment of the disease by physicians have contributed to these increases. Retrospective studies⁹ and prospective studies¹⁰ of asthma have been carried out in Alberta, Canada, which has had the highest mortality rate from asthma in Canada.¹ Near-fatal asthma (NFA) has been shown to share some features with fatal asthma,¹¹ but most studies of NFA^{12 13 14 15 16} have been retrospective and, generally, with patients from one ICU, although some have been prospective.^{11 17}

The criteria for a near-fatal exacerbation of asthma should include, as does this study, a stringent definition. Unlike studies of asthma mortality, patients with near-fatal exacerbations of asthma can be questioned directly regarding the characteristics of their disease and the events preceding the episode. These studies then can provide data that may help in identifying high-risk patients and situations. Interventions designed to prevent NFA are likely to be helpful in preventing asthma deaths.

The present study was population-based and included urban and rural areas in Alberta. The study compared patients who had experienced a near-fatal exacerbation of asthma (NFA patients or cases) with a group of patients who were treated for asthma in a hospital emergency department (ED [ED control subjects]) and a group of people in the community who had asthma (community control subjects [CCs]). The objective was to identify factors that distinguished the cases from the two control groups, particularly factors that have implications for clinical management.

	Materials and Methods

TOP Abstract Introduction Materials and Methods Results Discussion References

 
The study was an unmatched case-control study. Alberta has 2.5 million people and a unified public health-care system.

NFA patients were all individuals aged 5 to 50 years who had experienced a near-fatal exacerbation of asthma between August 1, 1994, and March 3, 1996, in Alberta, Canada. The age limits were chosen to minimize misdiagnosis. NFA patients were defined as having required assisted ventilation in an ICU in one of the cities or regional centers. Local clinicians, not the investigators, made the diagnosis of asthma and determined hospital admission policies and patient management. The study coordinator was notified of all hospital admissions for asthma (ie, 144 cases) to the participating ICU. Fifty-eight patients received assisted ventilation and were approached regarding the study by their physicians. All patients agreed to be contacted by the study coordinator, who obtained consent to mail the self-administered questionnaires. Completed questionnaires were received from 45 of 58 participants (77.6%).

Control Subjects
ED control subjects were aged 5 to 50 years and had been seen in the EDs at the University of Alberta Hospital in Edmonton or the Lethbridge Regional Hospital because they had sought treatment for asthma, and they also had received a hospital discharge diagnosis of asthma by a physician. ED control subjects were contacted by phone within 5 days of their ED visit and were invited to participate by completing a self-administered mailed questionnaire.

Phone follow-up for ED control subjects was undertaken to a maximum of nine times or until the participant declined participation. Mail reminders and additional questionnaires were provided to those who had misplaced the original questionnaire. Of the 390 potential cases, completed questionnaires were obtained from 297 (76.1%).

CCs between the ages of 5 and 50 years were identified through random-digit dialing according to a previously described procedure.¹⁸ From the > 7,000 random phone numbers, 393 people who told us that asthma had been diagnosed by a physician were identified, and of these 303 (77.1%) completed the questionnaire. The numbers of CCs recruited reflected the approximate 40:60 rural/urban composition of the province.

The proportion of control subjects (ED and CCs) to cases (NFA patients) was established as a minimum 3:1.

Questionnaire
The questionnaire was drafted by a group of physicians, epidemiologists, and asthma experts, was pretested in a pilot study of 35 patients with asthma, and took between 45 and 75 min to complete. The questionnaire requested information on demographics, age at first diagnosis of asthma, duration of symptoms before diagnosis, self-assessment of severity, symptom frequency, medication use (by broad category and specific names), history of allergy, exposure to asthma triggers, the asthma trigger for the index episode, and feelings of vulnerability.¹⁰ The asthma opinion survey,¹⁹ which evaluated feelings of vulnerability, was included. An example of a statement related to general health requiring a response is as follows: "I often worry about getting a serious disease such as cancer or a heart attack." An example of a statement about specific asthma vulnerability requiring a response is as follows: "I suppose that I will have some asthma attacks the rest of my life."

The study was approved by the appropriate research ethics boards. All participants were notified that the information was confidential and would not impact the type or quality of care. Parents were asked to complete questionnaires for participants who were aged 5 to 16 years.

Data Analysis
Questionnaires were reviewed manually for consistency and appropriate coding prior to data entry. All data were entered twice into the a computer program (SPSS-Data Entry; SPSS, Inc; Chicago, IL),²⁰ and discrepancies were resolved. The data were edited and were checked for appropriate codes and outliers, and for the normality of the distributions of continuous variables. The risk factors chosen for evaluation in bivariate and regression analysis were those based on the literature and on the expertise of the investigators.

Initial univariate and bivariate analyses examined case-control differences using ORs (ORs) and comparisons of means. When there was a gradient in the data, the Mantel-Haenszel test for linear trend was used. A computer program (Epi Info, version 6.0; Centers for Disease Control and Prevention; Atlanta, GA)²¹ was used for determining the ORs and confidence intervals (CIs) [Cornfield method] in the stratified analysis. A computer program (SPSS for Windows) was used for cross-tabulations and logistic regressions. The NFA patients (ie, cases) were compared to the control subjects for demographic variables, self-reported asthma severity, health-care utilization, medication use, and duration of disease. The most salient case-control differences were noted. Given the small number of cases, limited multivariate analyses were attempted. Within each group of related independent variables (eg, health-care utilization and asthma triggers), those variables exhibiting significant bivariate relationships (ie, p 0.05) were entered into a stepwise logistic regression analysis using forward selection. Finally, variables found to be significant in the logistic model within categories (eg, health-care utilization and asthma triggers) were entered into a summary logistic regression model to determine which remained significantly associated with case-control status. Logistic regression was preferred to discriminant function analysis because it requires fewer assumptions about the data (ie, multivariate normality of the independent variables and equal variances), and because it performs well even when the assumptions for discriminant function analysis are met.

	Results

TOP Abstract Introduction Materials and Methods Results Discussion References

 
The gender distribution was similar for the three groups (Table 1 ). However, 70% of the participants in the ED and NFA groups were under age 25 years, and 41.5% of ED participants were under age 15 years. The NFA group differed significantly from the ED group in age at diagnosis (p = 0.014), self-reported asthma severity (p = 0.001), and symptom frequency (p = 0.031). Those with NFA were more likely to self-report moderate-to-severe disease and more frequent symptoms than either of the control groups (Table 2 ). In a week when they were not having problems with their asthma, approximately half of the ED and the CC groups had no symptoms or had symptoms only with exercise, compared with 28.9% of the NFA group. Conversely, 33.4% of NFA group had symptoms more than three times per week, compared with 25.8% for the ED group and 21.8% for the CC group. When these variables were entered into a logistic regression comparing the ED group to the NFA group, age at diagnosis and disease severity remained significant at p 0.05. The NFA group differed significantly from the CC group in age at diagnosis (p = 0.001) and asthma severity (p = 0.007); in the logistic regression, age at diagnosis and severity remained significant (p 0.001).

The increased use of medication by patient self-report was more common among the NFA patients, with 79.5% describing increased medication use in the last 12 months, compared with < 60% of ED control subjects and only 38.5% of CCs (Table 3 ). The differences in the increased use of medication in the previous 12 months between the NFA group and the ED group was significant (OR, 2.60; 95% CI, 1.19 to 5.71), as was the frequency of night symptoms (OR, 2.96; 95% CI, 1.46 to 5.95). When the NFA group was compared to the ED group using logistic regression, the frequency of night symptoms remained significant at p 0.02. The following differences between the NFA group and the CC group were also significant: increased use of medication in the last 12 months (OR, 6.21; 95% CI, 2.87 to 13.3); frequency of symptoms in the morning (OR, 2.35; 95% CI, 1.23 to 4.52); and frequency of night symptoms (OR, 6.49; 95% CI, 3.18 to 13.16). When these variables were entered into a logistic regression, increased use of medication and frequency of night symptoms remained significant (p 0.001).

There were many environmental exposures in all groups, such as pets in the home and smoke, and many patients were current smokers as well. There were more smokers in the NFA group, but no significant differences were noted among the groups. There were no significant differences among the groups for asthma being triggered by smoke. Furthermore, there were no significant differences among groups for the proportion of patients whose mother or father had smoked during their childhood (² test). There were no significant differences among groups regarding the number of people in the household who smoked regularly or the number of hours per day the patients were exposed to smoke (one-way analysis of variance).

The use of medication was categorized as either never or occasional use compared to daily use. All forms of medication, including bronchodilators and inhaled steroids, were used significantly more frequently by participants in the NFA group than by those in the ED group (Table 5 ). The NFA group was more likely to received inhaled steroids daily (72.7%) and steroid pills daily or occasionally (77.2%), but inhaled steroids were not universally used.

The subjects in both the ED group and the NFA group were asked to identify the trigger for the episode that had led to an ED visit or had preceded an NFA episode. Only allergy or ingestion was significantly different, and they were present in six patients (13.3%) in the NFA group and two control subjects (1.0%) in the ED group (OR, 14.92; 95% CI, 2.51 to 153.8). Those in the NFA group were more apt to have reported that life had been stressful, to have reported a food allergy, and to have been to a bar, restaurant, or party. By comparison, the ED control subjects were more likely to have been engaged in physical activity while either recovering from a cold or having been exposed to someone with a cold. Similar proportions of NFA patients and ED control subjects reported that they had run out of medication or had been exposed to an allergen.

NFA patients also had the highest likelihood of having undergone previous intubation or cardiopulmonary resuscitation (30 patients; 71.4%) compared with 5.7% of ED control subjects (11 subjects) and 2.0% of CCs (6 CCs; p < 0.001). This history was by self-report and was not confirmed by inspecting previous records.

The vulnerability score was assessed among the groups, and it was noted that there were significantly higher scores on vulnerability for the NFA group compared to the CC group (p < 0.023). However, the differences between NFA and ED participants were not significant (p < 0.15) [Table 6 ].

	Discussion

TOP Abstract Introduction Materials and Methods Results Discussion References

 
This prospective, population-based, case-control study demonstrated differences among NFA patients, ED control subjects, and CCs. Medication use had increased significantly at > 12 months, and almost all participants had required a short burst of oral steroid therapy at some time, suggesting that there had been physician contact. The increased use of inhaled corticosteroid therapy among NFA patients (ie, 56.8% of NFA patients had used inhaled corticosteroids in the previous 12 months, a rate that rose to 72.7% in the previous month) suggests that there was a deterioration in asthma control before the NFA attack. The same trend in the use of inhaled corticosteroids, but at different levels, was noted in the ED control subjects, with 28.9% having used inhaled steroids in the previous 12 months, a rate that rose to a 45.9% rate of daily use in the previous month.

NFA patients had multiple markers of severity, including a high incidence of having undergone intubation and cardiopulmonary resuscitation, having previous hospital admissions, and having a strong history of allergic and atopic disease with life-threatening allergic reaction at some time in the past. They also had asthma symptoms on wakening and during the night.

There are no universally accepted criteria for NFA. All our subjects had undergone intubation. None had PaCO₂ levels of > 45 mm Hg and/or had experienced unconsciousness without intubation.¹¹ Some have defined all ICU admissions as NFA,^{6 21} but objective criteria are needed because of the wide variety of factors that may contribute to a decision to admit a patient to an ICU. Kallenbach et al¹⁶ included only patients who had experienced respiratory arrest or coma, and Wasserfallen et al²² included only those with very sudden onset of asthma. The criteria in both of these studies seemed to us to be too restrictive.

One advantage of our study was the use of a comprehensive questionnaire in a population-based study, avoiding the bias of individual institutional studies in which some population groups are not included and others are over-represented.^{17 23} In other contexts, the danger of generalizing from institutional studies rather than population studies has been pointed out.²⁴ Also, this study was carried out over a short time frame, and changes in practice over time likely did not affect the outcome.

In this population-based study, the number of cases (45) was comparable or greater than the number of cases in other studies of NFA. For example, an institutional study included 19 patients (cases) and 88 control subjects.¹⁷ Thus, patient numbers in NFA studies are likely to number < 50. More subjects may be recruited by involving other provinces or by increasing the duration of the study. These methods give rise to other problems. The relatively small number of NFA cases limits the power of the study to detect small differences, but many of the differences noted were large, and many ORs excluded the null value.

A limitation of the study was our use of recall to assess medication use. We went to great lengths to validate the data but found this difficult because of medication sharing, the use of multiple pharmacies, and biased consent to access pharmacy records. We had intended to validate the data, but most of those persons who gave permission to access the pharmacy records were in the NFA group.

We used self-assessment of asthma severity. Despite concerns about the self-assessment of asthma severity, some studies^{25 26 27 28} have shown that the results of self-assessment correlate with objective markers of severity, such as medication use and health-care utilization, and are a predictor of morbidity and mortality.

Some of the findings are of clinical importance. Despite the increased use of all types of asthma medication by the NFA group, including the observation that > 90% received therapy with inhaled steroids, the control of their asthma continued to deteriorate to a critical level. The circumstances preceding presentation to a hospital differed between the NFA and the ED groups. Those who experienced an NFA episode were twice as likely to have had that episode triggered by emotional events and to have reported being at a bar, party, or restaurant prior to the episode. By comparison, the ED group reported more often that an upper respiratory tract infection or physical activity precipitated their visit.

Smoking was common in all groups and was substantially more prevalent among the NFA group. Furthermore, smoking rates in the NFA group exceeded the provincial rate of 29.2%.²⁹ In addition, those patients with severe asthma did not seem to make any greater efforts than other patients to avoid smoke exposure.

Psychological issues in asthma exacerbation³⁰ have been noted, and we confirmed that feelings of vulnerability were common in the NFA group and that stress may be an important trigger. Stress was as perceived by the patient, but may affect the perception of the dyspnea.³⁰ Physician assessment of patients with severe asthma should include stress and vulnerability.

The issues raised by this study are clear. We confirm that patients who are admitted to an ICU with asthma have many markers of long-term severity and need intensive assessment and long-term follow-up. Effective treatment is likely to improve their quality of life and to allow participation in, for example, sports. Given their subsequent use of health-care resources,³¹ such an approach may reduce costs. However, there is considerable overlap in the risk factors among the three groups. Measures described in consensus reports such as attention to the severity of disease, the use of prophylactic drugs, and the avoidance of triggers are applicable for everyone with asthma.

	Acknowledgements

 
We thank the directors and staff of all ICUs in Alberta, Monica Ruff (Study Coordinator), and Diane Beauvais-Bishop and Christina Mauratt for patient secretarial help.

	Footnotes

 
Abbreviations: CC = community control subject; CI = confidence interval; ED = emergency department; NFA = near-fatal asthma; OR = odds ratio

This research was supported by the Herron Foundation Study of Childhood Asthma.

Received for publication March 15, 2001. Accepted for publication November 20, 2001.

	References

TOP Abstract Introduction Materials and Methods Results Discussion References

 

1. Mao, Y, Semenciw, R, Morrison, H, et al (1987) Increased rates of illness and death from asthma in Canada. Can Med Assoc J 137,620-624[Abstract]
2. Wilkins, I, Mao, Y (1993) Trends in rates of admission to hospital and death from asthma among children and young adults in Canada during the 1980’s. Can Med Assoc J 148,185-190[Abstract]
3. Robertson, CF, Rubinfield, AR, Bowes, G (1990) Deaths from asthma in Victoria: a 12 month study. Med J Aust 152,511-517[ISI][Medline]
4. Sears, MR, Rea, HH, Beaglehole, R, et al (1985) Asthma mortality in New Zealand: a two year national study. N Z Med J 89,271-275
5. . British Thoracic Association (1982) Death from asthma in two regions of England. BMJ 285,1251-1255
6. Rea, HH, Scragg, R, Jackson, R, et al (1986) A case-control study of deaths from asthma. Thorax 41,833-839[Abstract]
7. Bucknall, CE, Slack, R, Godley, CC, et al (1999) Scottish Confidential Inquiry into Asthma Deaths (SCIAD), 1994–6. Thorax 54,978-984[Abstract/Free Full Text]
8. Burr, ML, Davies, BH, Hoare, A, et al (1999) A confidential inquiry into asthma in Wales. Thorax 54,985-989[Abstract/Free Full Text]
9. Tough, SC, Green, FHY, Paul, JE, et al (1996) Sudden death from asthma in 108 children and young adults. J Asthma 33,279-285
10. Tough, SC, Hessel, PA, Ruff, M, et al (1998) Factors that distinguish those who die from asthma from community controls with asthma. J Asthma 32,657-665
11. Campbell, DA, McLennan, G, Coates, JR, et al (1994) A comparison of asthma deaths and near-fatal asthma attacks in South Australia. Eur Respir J 7,490-497[Abstract]
12. Lin, RY, Rehman, A (1995) Clinical characteristics of adult asthmatics requiring intubation. J Med 26,261-277[CrossRef][ISI][Medline]
13. Corn, B, Hamrung, G, Ellis, A, et al (1995) Patterns of asthma death and near-death in an inner-city tertiary care teaching hospital. J Asthma 32,405-412[ISI][Medline]
14. LeSon, S, Gershwin, ME (1996) Risk factors for asthmatic patients requiring intubation: III. Observations in young adults. J Asthma 33,27-35[ISI][Medline]
15. Newcomb, RW, Akhter, J (1988) Respiratory failure from asthma. Am J Dis Child 142,1041-1044[Abstract]
16. Kallenbach, J, Frankel, AH, Lapinsky, SE, et al (1993) Determinants of near fatality in acute severe asthma. Am J Med 95,265-272[CrossRef][ISI][Medline]
17. Turner, MO, Noertjojo, K, Vedal, S, et al (1998) Risk factors for near-fatal asthma: a case-control study in hospitalized patients with asthma. Am J Respir Crit Care Med 157,1804-1809[Abstract/Free Full Text]
18. Potthoff, RF (1994) Telephone sampling in epidemiology research: to reap the benefits, avoid the pitfalls. Am J Epidemiol 139,967-978[Abstract/Free Full Text]
19. Richards, JM, Jr, Dolce, JJ, Windsor, RA, et al (1989) Patient characteristics relevant to effective self-management: scales for assessing attitudes of adults toward asthma. J Asthma 26,99-108[ISI][Medline]
20. Norusis, MJ (1994) SPSS advanced statistics, version 6.1. SPSS Chicago, IL.
21. Dean, AG, Dean, JA, Coulombier, D, et al (1994) Epi Info, version 6; a word processing, database and statistics program for epidemiology on microcomputers. Centers for Disease Control and Prevention Atlanta, GA.
22. Wasserfallen, JB, Schaller, MD, Fechl, F, et al (1990) Sudden asphyxic asthma: a distinct entity. Am Rev Respir Dis 142,108-111[ISI][Medline]
23. Campbell, DA, Yellowlees, PM, McLennan, G, et al (1995) Psychiatric and medical features of near fatal asthma. Thorax 50,254-259[Abstract]
24. Layde, PM, Poraste, SK, Desbiens, N, et al (1996) Generalizability of clinical studies conducted at tertiary care medical centers: a population- based analysis. J Clin Epidemiol 49,835-841[CrossRef][ISI][Medline]
25. O’Conner, GT, Weiss, ST (1994) Clinical and symptom measures. Am J Respir Crit Care Med 149,S21-S28
26. van der Windt, DA, Nagelkerke, AF, Bouter, LM, et al (1994) Clinical scores for acute asthma in pre-school children: a review of the literature. J Clin Epidemiol 47,635-646[CrossRef][ISI][Medline]
27. Tough, SC, Green, FY, Paul, JE, et al (1996) Sudden death from asthma in 108 children and young adults. J Asthma 33,179-188[ISI][Medline]
28. Dales, RE, Schweitzer, I, Kerr, P, et al (1995) Risk factors for recurrent emergency department visits for asthma. Thorax 50,520-524[Abstract]
29. . Statistics Canada (1995) Percentage of smokers in the population, 1994–1995: national population health survey. Statistics Canada Ottawa, Canada. Catalogue No. 82–221-XDF
30. Boulet, LP, Deschesnes, F, Turcotte, H, et al (1991) Near-fatal asthma: clinical and physiologic features, perception of bronchoconstriction and psychologic profile. J Allergy Clin Immunol 88,838-846[CrossRef][ISI][Medline]
31. Kesten, S, Chew, R, Hanania, N (1995) Health-care utilization after near-fatal asthma. Chest 107,1564-1569[Abstract/Free Full Text]

This article has been cited by other articles:

				N.C. Thomson, R. Chaudhuri, and E. Livingston Asthma and cigarette smoking Eur. Respir. J., November 1, 2004; 24(5): 822 - 833. [Abstract] [Full Text] [PDF]

				A. F. Gelb, A. Schein, E. Nussbaum, C. M. Shinar, Y. Aelony, H. Aharonian, and N. Zamel Risk Factors for Near-Fatal Asthma Chest, October 1, 2004; 126(4): 1138 - 1146. [Abstract] [Full Text] [PDF]

				E. Barreiro, J. Gea, C. Sanjuas, R. Marcos, J. Broquetas, and J. Milic-Emili Dyspnoea at rest and at the end of different exercises in patients with near-fatal asthma Eur. Respir. J., August 1, 2004; 24(2): 219 - 225. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Free Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Article Archive Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (32) Citing Articles via Google Scholar Google Scholar Articles by Mitchell, I. Articles by Hessel, P. A. Search for Related Content PubMed PubMed Citation Articles by Mitchell, I. Articles by Hessel, P. A.