Asthma is a common disease, occurring in 5% to 7% of children in the United States.1 Its prevalence and often mild clinical course have led to a widely held belief that patients do not die from asthma. Osier even stated in the early 190Os that "the asthmatic pants into old age."2 Actually, death due to asthma has heen known to occur since the early days of medicine, with the problem described in medical writing since the 12th century.3
Rackemann's work in the 1940s and 1950s ^acquainted the medical community with the reality of deaths due to asthma.4'5 In addition, he was the first author to note that death due to asthma could occur in children.'' Death in children, especially adolescents, has become a particular concern since the "epidemics" of asthma deaths that occurred in England and Wales in the 1960s6 and in New Zealand in the 1970s.7 Adolescent deaths have increased, as deaths due to asthma in younger children have decreased dramatically. Deaths in adolescents most often occur outside the hospital and therefore have not been affected by improvements in hospital care.
With the realization that death can result from asthma, research has focused on identifying risk factors and preventing deaths. The purposes of this article are to describe how patients can place themselves at high risk for dying by their behavior, to review features of physician care and the medical system that may place patients at risk, and to recommend an approach to individual patients that can effect a more favorable outcome. This approach to patients can also reduce morbidity from asthma, repeated emergency room visits, and hospitalizations.
CHARACTERISTICS OF PATIENTS WHO DIED
Although patients of any age, sex, and race can die from asthma, there are differences for rates of death by race and sex.8,9 In the United States, blacks have a higher rate of death from asthma (2.1 per 100 000) than whites (1.2),8 Interestingly, this increased rate in blacks is mirrored by an increased rate in nonwhites (Maori and Polynesians) in New Zealand in the recent asthma death epidemic there. 7 The death rate is somewhat higher for females (1.5) than males (1.2), although this difference has not been as constant over years as that between races.
The foremost risk factor for death from asthma is severe disease. Although death is quite rare when considering the entire population of asthmatics, approximately 1% to 2% of severe asthmatics will die as a result of their disease.10,11 The asthma observed in these patients is clinically diverse; however, two general groups are apparent. The first group had a near death episode requiring resuscitation. Sudden severe asthma of this type may or may not be accompanied by ongoing severe symptoms. Twenty to thirty percent of patients who have died of asthma have had previous threatening events.12 The second group is those individuals who require cor t icostero ids for control of chronic severe symptoms.13 These patients may also have a history of hypoxic seizures, respiratory failure requiring ventilation, severe nighttime wheezing, or wide rapid fluctuations in pulmonary functions from normal to abnormal. Some patients do not fit into either classification and had extremely poor control of their asthma in the month before their death.14 Many were recently hospitalized.15
Poor self-care, another major risk factor,13 is typified by missed appointments, incorrect use of medications, and poor knowledge of asthma. Disruption in the physician-patient relationship may result in disregard of the physician's instructions. This is particularly relevant when se If- initiated reduction of steroids occurs. Also, inappropriate response to symptoms by either disregarding wheezing or using asthma to manipulate others increases the risk.13
Psychological problems have often been noted in both the children and their families. In a study by Strunk et al,13 thorough evaluations of physiologic and psychological features were available for a group of 21 adolescents who died of asthma. Case controls were matched for age, sex, and severity of disease. Records for the 21 cases and 21 controls were evaluated for 43 physiological and 14 psychological criteria. Of these 57 variables, 14 were present in significantly different percentages in the group of children who died of asthma when compared with the controls. Eleven of the 14 differentiating criteria reflected the psychological adaptation of the child or the child's family with only three characterizing the physiologic status of the child or the severity of the asthma. The psychological variables were grouped into three clusters:
1. Self-management problems were apparent in three separate variables (disregard of perceived asthma symptoms, self-care inappropriate for the age of the child, and use of asthma to manipulate parents or teachers);
2. Poor family support for asthma management was apparent in difficulties in family cooperation with the medical plan (parent-staff conflict), more general examples of parent-child conflict (eg, fights over asthma treatment), and overall family dysfunction or crisis (eg, physical abuse and neglect, alcoholism, intense marital conflict, mobile family, fatalistic attitude, and inability to cope with financial difficulties);
3. Psychological problems of the children were apparent from a psychiatric diagnosis, observations of depressive affect and excessive sensitivity to loss or separation.
Wheezing with stress was noted more frequently in children who died than in the controls. Furthermore, discriminant analyses indicated that the physiologic or disease severity variables alone did not identify the patients who died unless there were also psychological difficulties present.
The importance of the psychological issues was also emphasized by a case-control study conducted by Rea et al15 in New Zealand in 1982. Patients (aged 15 to 60) who died from asthma were compared with those hospitalized for asthma in the same period. A history of life-threatening asthma attacks and psychological problems distinguished the two groups. The psychological problems included recent bereavement, unemployment, alcoholism, depression, and personality disorders.
The final patient characteristic mentioned by some authors is severe atopy. Investigators from the United Kingdom noted a predominance of deaths in the spring and fall when patients are atopic.16 Other authors have noted a history of a large allergen exposure immediately before the deaths of some patients; these patients appeared to die abruptly with rapid onset of severe bronchospasm.17'18
These characteristics are translated into several issues that are apparent on the day of death. Many patients and their parents appear to have been complacent, thinking that the current attack would resolve just as the others had. Apparently, some of the parents had been told that death could not happen. Second, poor assessment of the severity of the attack by the patient and parent resulted in delays in contacting the physician. In many cases, reasons for delays are use of β2-adrenergic agents, which provide relief that is progressively less in both magnitude and duration. Third, the attack was poorly treated by the patients and parent, not only in the overuse of β2-agonists, but the underuse of corticosteroide. In addition, many patients seemed to be confused about what treatment to use. Finally, some patients have had a wish to die or a premonition of death.
The reasons listed as the basis of final outcome can be derived directly from the issues apparent on the day of death: the course of the attack appears rapid, often because early signs of worsening were not recognized, there is poor access to care, and the patients usually arrive late for the care.
The results of the case reports and the case-control studies suggest that patients with severe disease, particularly life-threatening attacks, are at high risk of death. Disease that is out of control, even in someone who otherwise has more mild illness, is probably an important factor as well. Psychological factors are frequently mentioned, and many seem to focus on issues that may interfere with the delivery of care, especially in an emergent situation.
PHYSICIAN CARE AND THE MEDICAL SUPPORT SYSTEM
Physicians may place patients at risk by failing to diagnose asthma or failing to recognize severe chronic disease (Table 1). Rea et al found decreased use of objective measures of pulmonary function in patients, ie, pulmonary function tests or peak flow measurements.19 Shortcomings in supervision and education may result in repeated failure to contact physicians during exacerbations, to recognize severity of deteriorations, or to follow up with medical care.
Acute attacks require prompt attention. Studies have found that some physicians underestimate the severity of an attack, delaying admission or resulting in the inappropriate use of medications.20 Deaths in hospitals were almost all associated with poor monitoring, inappropriate drug use (especially corticosteroids), and use of sedatives. A case-control study reported by Eason and Markowe in 1987 of hospital care in the Northeast Thames region of the United Kingdom21 strongly supports the importance of quality hospital cate in the outcome of asthma in some patients. Monitoring of the clinical course of the illness, especially with arterial blood gases, was more often déficient in fatal cases of asthma than in the patients who survived an episode of status asthmaticus.
Although some deaths cannot be prevented, common problems of underrecognition of asthma severity and delays in seeking care can be approached with current methods of care. Thorough evaluations should include some assessment of the severity of attacks and rapidity of onset.
A patient approach designed to prevent deaths has three components: educating both patients and parents about asthma, establishing a medical regimen to control the asthma, and scoring for high-risk status. When high-risk status is ptesent, there should also be special planning.
Patient education is more than simply having the patient and family take an asthma education course. Although these courses are valuable, they are not all that is necessary. What is needed is ongoing follow-up, reinforcement of asthma education and, more importantly, its use. Patient technique with metered-dose inhalers should be checked at each visit, and guidelines for use of the drugs and potential adverse reactions (including possible problems of overuse) should be reviewed at regular intervals. Ongoing education intervention includes encouraging compliance, even when overall improvement is evident, by teaching each patient that asthma waxes and wanes over time but seldom completely disappears. Counseling should include the importance of prompt treatment of symptoms and physician contact to obtain supervision.
When to Refer to an Asthma Specialist and Questions to be Answered
Like asthma education, the importance of establishing effective maintenance therapy is obvious. The effectiveness of asthma therapy must be confirmed regularly for all patients by both pulmonary function testing at each patient visit and by confirming that the regimen allows ongoing participation in activities of the patient's choice and minimizes absenteeism from work or school. Since many patients assume that ongoing symptoms and inactivity due to exerciseinduced asthma is their burden in life, simply asking if the patient is having problems with his asthma is insufficient. Detailed questions regarding the type and extent of exercise and school or work absenteeism must be asked regularly. Many authors recommend the use of mini peak flow meters at home to aid in detecting increasing airway obstruction so that appropriate therapy can be initiated early in the course of the asthma.
Patients should be scored for high-risk status22 (Table 2) and such a designation should initiate more frequent follow-up and special planning to treat features that result in high-risk status. 23 There are at least two types of patients at risk: 1) patients with a single life- threatening episode of asthma, regardless of the underlying severity of the disease on an ongoing basis, even if all other circumstances are optimal; and 2) severe asthma of any type, especially if there are accompanying psychosocial problems that could yield either poor communication, poor compliance, or hopelessness about the chronic illness. A list of possible psychological problems is presented in Table 3.
Questions to Ask to Identify a High-Risk Child
High-Risk Psychological Variables
Special planning is required for the high-risk patient. These patients should be followed more closely, both in frequency and intensity. Features of their case that result in high-risk status should be treated specifically, for example, more intensive education for poor self-care and consideration of psychiatric referral for help in dealing with the psychological problems, especially for family dysfunction that could interfere with compliance with the medical regimen. In developing a plan for an overall approach to these patients, it is often useful to identify an advocate close to the patient who can watch out for the patient's welfare and to involve community resources, including a school. Finally, physicians should notify patients about their high-risk status. The content of this discussion will vary widely depending on the developmental stage and the capabilities of the child and family to deal with the issue. At a minimum, patients should be told that they have severe disease and that if they do not pay attention and communicate well with a physician early in the course of an attack they may have serious problems. In addition, physicians should not support the myth that asthma can never be fetal.
Finally, plans for approaching acute problems should be drawn up for each patient and should include:
* Criteria for initiating the crisis plan with clear guidelines on how much additional treatment the patient may use before contacting the physician. Typical rules might be contact after two extra treatments are needed in a 24-hour period or if more than two treatments are needed in less than 4 hours.
* What to do if the physician cannot be contacted or until the physician calls back. Often this includes taking a predesignated dose of prednisone, usually 1 to 2 mg/kg depending on the course of previous attacks and the amount of the drug required.
* Planning for care during vacations or other periods away from the usual sources of medical care.
* Special planning for patients in the event of severe asthma. This might include identification of an emergency room and paramedic squad and communication about care that might be necessary if the patient should present to them.
1. Gergen PJ, Mullay DI. Evans R: National survey of prevalence of asthma among children in the United Stana, I976-198C. Pediatrics 1988; 81:1-7.
2. Osier W: The Principles and Procace of Medicine. Edinburgh, Young J Pentland, 1892, p 498.
3. Siegel SC History of asthma deaths from antiquity- 3 ABergj CJm Immurici 1987; 80:458-462.
4. Raclcemann FM: Deaths from asthma. } Allergy 1944; 15:249-258.
5. Rackemann FM, Edwards MC: Asthma in children. A follow up study of 688 patients after an interval of ZO years. N Engl ] Mtd 1952; 246:815-823. 858-863.
6. SpeuerFE. DoIIR1 Heat P: Observations on recent increase in mortal ity from asthma. Br Med J 1968; 1:335-339.
7. Jackson RT, Beaglehole R1 Rea HH, et al: Mortality from asthma: ? new epidemic in New Zealand. Br Med} 198Z: 285:771-774.
8. Sly RM: Increases in death from asthma. Annoii of Asdana 1964; 53:20-25.
9. Evans R, Mullally Dl. Wilson RW, et al: National trends in the morbidity and mortality of asthma in the U.S. Prevalence, hospital nation and death From asthma over two decades; 1965-1984. Chesi 1987; 91:655.
10. Mellis CM. Phelan PD: Asthma deaths in children: A continuing problem. Thorax 1977; 32:29-34.
11. Blair H: Natural history of childhood asthma. Arch Du Cfiiü 1977; 52:613-619.
12. Sears MR. Rea HH. Beaglehole R. et al: Asthma mortality in New Zealand: A two year national study. NZ Med J 198S; 98:271-275.
11. Strunk RC. Mrawk CA, Fuhrmann GSW, et al: Physiologic and psychological characteristics associated with deaths due to asthma in childhood. iAMA 1985; 254:1193-1198.
15. Attaway NJ. Birkhead GS, Townsend MC, el ah Investigation of a cluster of asthma deaths in teenagers. } Allergy CIm Jmmunol 1988; 81:306.
15. Rea HH, Scragg RR, Jackson R. et al: A case controlled study of deaths from asthma.
Thorax 1986; 41:83 3-8Î9.
16. British Thoracic Society: Comparison of aropic and non-atopic patients dying of asthma. Br ] Du Chest 1987; 81:30-34.
17. O'Hollaren MT, Sachs MI, YungingerJW. et al: Alternaría sensitivity as a possible risk (actor for respiratory arrest in children with asthma. ] Allergy CIm !nominal 1986; 77:199.
18. O'Hollaren MT, Sachs Ml, O'Connell EJ, et al: Allergen exposure as a possible precipitating factor for respiratory attest in young adults with asthma. J Allergy CIm lirammol 1988; 80:256.
19. Rea HH, Sears MR, Beaglehole R, et al: Lessons from the national asthma mortality study: Circumstances surrounding death. NZ Med J 1987; 100:10-13.
20. British Thoracic Association: Death from asthma m two regions of England. BrMeJJ 1982; 285:1253-1255.
21. Eason J, MarkowHLJ: Controlied investigation of deaths from asthma in hospitals in the North East Thames region. Br Med J i987; 294:1255-1258.
22. Strunk RC: Summary of workshop discussion. Workshop on the identification of the fatality-prone patient with asthma. The asthma mortality task force. J Allergy Clin immimol 1987; 80:455-457.
23. Srrunk RC: Deaths from asthma in childhood: Patterns before and after professional intervention, ftdiutr Aslhma Allergy Imrnioiol 1987; 1:5-13.
When to Refer to an Asthma Specialist and Questions to be Answered
Questions to Ask to Identify a High-Risk Child
High-Risk Psychological Variables