Assessment of Practical Knowledge of Self-Management of Acute Asthma

Assessment of Practical Knowledge of Self-Management of Acute Asthma* John Kolbe, MBBS; Marina Vamos, MB ChB; Frances James, PhD, Gail Elkind, PhD; and Jeffrey Garrett, MB ChB Aims: To develop an instrument for the measurement of, and to determine the level of, practical knowledge of self-management of acute asthma. Methods: Eighty patients with moderate/severe asthma attending a hospital-based asthma clinic responded to an interviewer-administered questionnaire. Subjects were asked to describe the action they would take in response to each of two hypothetical evolving attacks: (1) one of gradually increasing sevmity and (2) the other developing rapidly. Responses were scored according to the appropriateness of actions taken relevant to the stage of the attack. Transcripts of the responses were scored independently by three of the investigators according to a system based on Thoracic Society of Australia and New Zealand (TSANZ) and British Thoracic Society (BTS) consensus statements on asthma management. A 25-point scale was used on which 0 represented a total lack of appropriate responses and a score of 25 was an optimal response. Results: Interrater and intrarater reliability were excellent. Mean (± SD) scores for the slow and rapid onset attacks were 12.8±4.0 and 13.9 ± 4.8, respectively. The scores for the two scenarios were predicted by each other (p=0.002) and by the interviewer's rating of asthma management knowledge (p=0.0004, p=0.0001), but not by age, sex, race, previous asthma morbidity, depression, or anxiety. In both scenarios, most patients indicated that they would increase inhaled Jl-agonist (85% for slow-onset scenarios and 94% for rapid-onset scenarios, respectively) and use their action plan and/or seek urgent medical advice at an appropriate time (74% and 70%). Although some would measure peak expiratory flow (PEF) initially (54% and 30%),

Jn asthma, the relationships among education, knowl-

edge, illness behavior, and morbidity remain unclear. Traditionally, asthma education has focused on teaching patients the pathophysiology of asthma, idenFor editorial comment see page 9 *From th e Der.artments of Respirato1y Medicine (Drs. Kolbe and Garrett), Psychiatry (Drs. Vamos and James), and Medicine (Dr. Kolbe), Green Lane Hospital and Umversity of Auckland School of Medicine, Auckland, New Zealand. Supported b y agrant from the Health Hesearch Council of New Zelliand. Manuscript received August 23, 1994; revision accepted July 17, 1995. 86

only a minority would continue to monitor PEF in the context of worsening acute asthma (30% and 24%). When a severe life-threatening situation was described, only 50% and 64%, respectively, indicated that they would call emergency services. Conclusions: Scenarios describing hypothetical asthma attacks are a useful and reproducible method of assessing practical knowledge of self-management of acute asthma. Patients presented with scenarios frequently made errors in their hypothetical responses. The errors made with scenarios, which parallel errors reported in real clinical situations, occurred despite the fact that this patient population had received considerable education and training about how to manage asthma. Most indicated they would not monitor PEF even in an exacerbation of asthma and would not call emergency services despite life-threatening asthma. These scenarios may allow us to explore the gap between knowledge about treatment and actual practice, and perhaps to help close that gap and thus reduce asthma morbidity and mortality. (CHEST 1996; 109:86-90)

BTS=British Thoracic Society; ED=emergency department; EMS=emergency medical services; GP=general practitioner; PEFR=peak expiratory flow rate; PFM=peak flow monitoring; TSANZ= Thoracic Society of Australia and New Zealand

Key Words: asthma; education; psychological factors; selfmanage ment

tificalion and modification of higgers, and the mechanism of action of drugs. While such programs have demonstrated effects on measures of asthma knowledge, the effects on traditional indexes of asthma morbidity such as hospital admissions, emergency department (ED) visits, and urgent physician visits are modest. 1•2 To obtain maximum impact, asthma education needs to be relevant, realistic, and repeated. 3 Instruction in the more practical aspects of self-monitoring of asthma, regular use of preventive medication, recognition and assessment of the severity of an asthma attack, the pharmacologic and nonpharmacologic means of aborting the attack, and when and how to Clinical Investigations

seek medical help if such strategies are not effective would seem intuitively more likely to influence morbidity. Future studies would be facilitated by a reliable and valid measure of asthma self-management knowledge. One method of assessing knowledge of the practical aspects of the management of acute asthma is to use questionnaires describing hypothetical attacks of asthma (scenarios) 2A· 5 that are appropriate for the hypothesis being addressed. Because of the increase in asthma deaths 6•7 and severe life-threatening attacks, 8 scenarios describing the development of severe attacks and thus assessing a patient's ability to deal with them might be particularly valuable. We therefore undertook a study (1) to test the feasibility and reliability of scenarios as a means of assessing practical knowledge of asthma self-management, (2) to determine the level of self-management knowledge of acute asthma in a group of moderate/ severe asthmatics attending a hospital-based asthma clinic, and (3) to determine whether any clinical and psychological factors correlated with scenario scores. METHODS

Patients

Subjects aged 14 years and older and diagnosed as having asthma, on the basis of demonstrable reversible airflow obstruction, were recruited from a hospital-based asthma clinic. The lower age limit is the age at which patients are transferred to adult clinics and are considered to be of sufficient maturity to take responsibility for their own health. The asthma clinic is located at a major public hospital that is the regional center for respiratory medicine. Although general practitioners may refer patients to the clinic, most referrals result from recurrent hospital admissions or ED attendances. Since 1982, all patients presenting with severe lifethreatening asthma and admitted to an ICU have been automatically referred to the asthma clinic. The clinic is staffed by respiratory physicians, a full-time asthma nurse educator who reviews patients at each visit and is available for follow-up in the community, a respiratory physiotherapist, and a clinical psychologist. Individualized written plans for self-management of acute attacks (action plans) have been an important part of patient education at the clinic for well over a decade. These instruct the patient on how to identifY an impending attack, what action to take, and when and how to seek medical help if these actions fail or the attack becomes severe. Peak flow monitoring (PFM) has been an integral part of this process. Interoieu;s and Questionnaires

The two hypothetical attacks of Sibbald4 were modified for administration by an interviewer. The first was an attack of increasing severity over 7 days (slow onset) while the second described an attack that developed over 1 h (rapid onset) (Appendix 1). Both scenarios ended with the subject "experiencing" a severe attack such that he/she was so wheezy and short of breath as to be unable to speak or rise from a chair. At three stages during each of the scenarios, subjects were asked to describe what action they would normally undertake if they were actually experiencing such symptoms. The scenarios, although based on those of Sibbald, 4 differed in several important ways: (1) introductory statements explaining the nature of the scenarios were included; (2) changes in wording were made to reduce ambiguity; (3) they were interviewer administered

(as opposed to self-administered); (4) questions were open ended such that specific management strategies were not suggested; a single prompt at each stage was used; and (5) scenario responses were taped and verbatim transcripts were scored independently by three of the investigators (for interrater reliability). A scoring system for each scenario was devised based on Thoracic Society of Australia and New Zealand (TSANZ) 9 and British Thoracic Society (BTS) 10·11 consensus statements on the management of acute asthma (Appendix 2). There was a 25-point scale on which 0 represented a total lack of appropriate responses and a score of 25 was an optimal response. Respiratory physicians associated with the study considered a score of 15 or greater would indicate a satisfactory level of asthma self-management knowledge. Subject responses were scored according to the appropriateness of actions taken relevant to the stage of the attack, eg, seeking medical advice was considered inappropriate in the initial stage of the attack and therefore did not score points, whereas at a later stage, when appropriate, it did. Scores were weighted for strategies considered most important in aborting an attack or to be potentially lifesaving. Negative scoring was not used. A sample of 20 responses was rescored by the investigators after a period of several months to test intrarater reliability. Other measures included as part of the interview were as follows: patient demographics; level of anxiety and depression (from the Hospital Anxiety and Depression Scale12 that is specific for distress in physically ill subjects); social support (measured by a modification of the scale of O'Reilly and Thomas 13 that was designed to evaluate social support in patients with cardiac disease; this included an assessment of general support and disease-specific supp01t, both day to day and during acute attacks ); and asthma morbidity over the previous 12 months as assessed by number of ICU admissions, hospital admissions, ED visits, and courses of oral corticosteroid therapy. This information was extracted from the prospectively compiled asthma clinic database. A single interviewer (F.J.) was used throughout. A global rating (five-point scale) of the subject's asthma self-management ability was recorded by the interviewer immediately at the conclusion of the interview. Also, a global rating (five-point scale) of the subject's overall asthma severity was recorded by each of three respiratory physicians after review of the medical record. Ethics

All subjects gave written informed consent to participate in the study that was approved by the Auckland Area Health Board Ethics Committee. Statistical Analysis

Data are expressed as mean ±SD. Interrater and intrarater reliability was assessed by percentage agreement for each behavior type. Behaviors were coded as occurring if two of the three raters recorded that behavior. General linear modeling was used to evaluate which of the demographic and morbidity variables predicted the slow and rapid onset scenario scores. A pvalue ofless than 0.05 was regarded as significant. RESULTS

Subjects Eighty subjects aged between 14 and 76 years (mean=41.2 years) were recruited from the asthma clinic. Forty-four percent were male. Eighty-nine percent were European and 9% were Maori or Pacific Islanders. Subjects had attended the clinic an average of 9.8 occasions, 3.8 in the previous year. On each occasion, they had received instruction from a physician CHEST /109/1/ JANUARY, 1996

87

Table !-Summary of Responses (%) to Slow-Onset Attack Scenarios*

Onset Progression Day 0 Inhaled 13-agonist Use of peak flowmeter Increase inhaled steroid

~ ~

@]

59

Life-

Life-

Threatening Attack Day 7

32

Inhale d 13-agonist

30

Use of peak flowm ete r

14

Increase inhaled steroids

Increase oral corticosteroid

~

5

Seek medical help

~

15

Call emergency services

Table 2-Summary of Responses(%) to Rapid-Onset Attack Scenarios*

~

Onset Oh

Progression

Threatening Attack 1h

~ ~ ~

50

33

24

~

Initiate oral corticosteroids/ medical advice

14

Call e merge ncy services

*In each case, the proportion of subjects indicating they would normally take a certain action at a particular time in the attack is shown. Boxes indicate the stage of an attack when it is considered most appropriate to initiate such an action .

*In each case, the proportion of subjects indicating they would normally take a certain action at a particular time in the attack is shown. Boxes indicate the stage of an attack when it is considered most appropriate to initiate such an action.

and the nurse educator and all had received a written action plan, a peak flowmeter, and instructions on PFM. Virtually all had moderate/severe asthma; all were taking inhaled cmticosteroids, 73% had required oral corticosteroids in the last year, 29% had been admitted to hospital in the previous year for acute asthma, 70% had previously experienced an attack of similar severity to that described in the slow-onset scenarios and 59% similar to that described in the rapid-onset scenario, and 34% were rated by respiratory physicians as having severe or very severe asthma.

meters; substantially fewer would check response to [3-agonists or continue PFM during the worsening attack. Only 42.5% indicated that they would have increased the dose of inhaled corticosteroids, an integral component of most current action plans. Only half indicated they would use emergency services during the week, despite the development of life-threatening severity. In the rapid-onset attack scenario, again the vast majority indicated that they would initially use inhaled [3-agonists, but substantially fewer indicated that they would use a peak flowmeter (16%) or increase inhaled steroids (15%). About 30% would not initiate oral corticosteroid therapy or seek medical help, despite a rapidly developing attack, so severe that the subject was so wheezy and short of breath that he or she had difficulty walking from one room to another. Just under 40% indicated they would not use emergency services, despite the scenario describing an attack in which the subject was unable to speak or get out of a chair.

Interview and Questionnaire

Interrater and intrarater reliability was excellent with concordance rates of more than 80% for virtually all items. Scores (out of a possible 25) for the two scenarios were as follows: slow-onset attack=l2.8:±:4.0; and rapid-onset attacbl3.9:±:4.8. The scores for the slow-onset attack were predicted by the score for the rapid-onset attack (p=0.0024) and the interviewer overall rating of management ability (p=0.0004) . Similarly, the scores for the rapid-onset attack were predicted by the score for the slow-onset attack (p=0.0026) and the interviewer overall rating of management ability (p=0.0001). Neither scenario score was predicted by age, sex, race, anxiety score, depression score, level of social support (general or speciflc), any of the indexes of asthma morbidity, or whether the subject had previously experienced an attack of similar type and sevelity. When the scores of the scenarios were examined in more detail, the responses can be summarized as in Tables 1 and 2. In the slow-onset attack scenario, most subjects indicated they would initially use inhaled [3-agonists and many would continue to use these agents. About half would initially use their peak flow88

DISCUSSION

We have developed scenarios to assess practical knowledge of asthma self-management. Despite considerable education, our patients were surprisingly poor at applying knowledge to action in this scenario situation. Furthermore, the deficiencies in conceptional plans reflected in the suboptimal responses to scenarios correspond surprisingly well with the types of self-management deficiencies that are encountered in actual clinical medicine. The data suggest that scenarios may be a tool that can be used to assess the true efficacy of teaching, particularly in high-risk patients who are justifiably targeted in asthma education programs.3 However, it is still necessary to demonstrate Clinical Investigations

that different educational initiatives can lead to different and improved responses to the scenarios and that this translates into changes in patient behavior, which in time, lead to improvement in self-management in real clinical situations. The scoring system develo~ed for this study was based on TSANZ9 and BTS 10· 1 guidelines. Although it is acknowledged that some of the advice contained \vithin these guidelines does not have asolid scientific basis, it does r epresent the consensus view of a large body of respiratory physicians and forms the basis of much of the educational material and information regarding the management of an acute attack given to patients at the clinic from which the subjects were recruited. There was excellent interrater and intrarater consistency in scenario response scores. Furthermore, the strong correlation of scenario scores \vith the interviewer's global rating of the subject's ability to recognize and respond appropriately to worsening asthma adds to the validity of scenarios as a m ethod of assessing practical self-management of severe attacks. The responses to the scenarios were, on the whole, disappointing. Patients attending the clinic are routinely given action plans that are based on PFM, but most subjects indicated that they would not undertake PFM, e ven during exacerbations of asthma. This is consistent \vith the results of Garrett et al2 who showed that in a multiracial working class neighborhood, only 16% of patients who possessed p eak flowm eters perform ed daily PFM, only 8% considered PFM at early stages of an asthma attack, and 26% rarely or never used their peak flowmeter. Perhaps a r eassessment of the "reliance" placed on PFM in advice to patients and contained in a variety of action plans currently available is required. If most patients do not use their peak flowmeter, e ven after extensive instruction, there seems little point in basing management on these measurements. Just as action plans need to be individualized, so too does the decision regarding the supply and method of use of a p eak flowmeter. More disturbingly, a significant proportion indicated that they would not call emergency services, despite what was clearly apotentially life-threatening situation. The proportion not seeking such help was higher in the slow-onset attack scenario, supporting the contention that slowly developing attacks over a number of days are particularly prone to being inadequately managed and are therefore potentially more hazardous. This is of particular concern, as most of those attending such a clinic have already been identified as "at risk" because of their high asthma morbidity, often \vith a history of a previous life-threatening attac0 in the recent past. In addition, none of the demographic or psychological parameters measured predicted the self-manage-

ment knowledge scores. Measurement of other factors that might influence the level of self-management knowledge, such as social and economic factors , patient's educational attainment, emotional maladjustment, attitudes to the disease and its management, relationship \vith health professionals, etc, was beyond the scope of this pilot study. However, clear identification of the factors that influence asthma knowledge \vill have important implications for future strategies aimed at improving self-management abilities and redncing asthma morbidity and mortality. In conclusion, this study supports the use of scenarios to measure asthma knowledge and, specifically, modifications of the Sibbald4 scenarios to assess practical knowledge of self-management of severe asthma. Such instruments \vill facilitate studies that are required to define the relationship between knowledge/ self-management skills, actual behavior, and morbidity.3 In doing so, these instruments should help define efficacious behaviors. This study also shows that despite current education strategies, many patients \vith moderate to severe asthma may make serious errors in the self-management of an attack and may not appropriately use emergency medical services, even during a potentially life-threatening attack of asthma. ACKNOWLE DGMENT: We thank the staff of Asthma Clinic for their patience and tolerance, all the subjects who participated for their time and effort, and Kathy Gallimore, Josephme Ratnasab
1

Asthma Knowledge Hypothetical Asthma Attacks (Based o n Sibbald 4) Now I am oging to ask you some questions about the kind of experiences that some people with asthma may have and I would like you t o talk about the sorts of things you would do if they were to happen t o you. I will begin b y describing to you a s hort scene about a particular situation. Then I will ask you some questions and then go on to the next scene. Some of your answers may cover all tbe stages. I must read you every scene so if that happens just repeat anything that relates to that p articular scene. I will tape our conversation because writing slows things down. If th ere is a ynthing you do not understand, tell me and I w ill go over it again. A. Slow-Onset Attack (D ays l to 3 ) For the last 2 days you have been f eeling a little more w heezy and breathless th an u sual, but not enough t ointerfere ~~th your everyday acti~ti es . Last night you woke up once because of asthma, but were able to get to sleep again easily. This morning you again woke up feeling more wheezy and breathless than usual. What would you do? Prompt(i ) Would you do anything e sl e? (or anything after that?) (ii ) Would you t ake a ny xe tra medication? (Days 4 to 5) It is now for 3 days that you have been m ore hs ort of breath and wheezy. Your breathing got slightly worse over the next 2 days and you f ound it increasingly difficult to get on with your everyday acCHEST / 109 / 1 / JANUARY, 1996

89

tivities. Last night you were wakened twice because of asthma and found it difficult to get back to sleep. Today you wake earlier than usual and are feeling very wheezy and breathless. What would you do? Prompt(i) Would you do anything e lse? (or anything else after that?) (ii) Would you take any extra medi cations? (Days 6 to 7) It is now 5 days that you have been more short of breath and wheezy. Your wheezing and breathlessness got worse over the next day. Last night you wakened three times because o fasthma and the last time you could not get back to sleep. It is now morning and you are so wheezy and breathl ess that you Hnd it diHicult to speak or walk across the room. What would you do? Has an)thing like this ever happened to you? (Yes/No)

B. Rapid-Onset Attack (Stmt) You woke this morning feeling pe1fectly well and spent the day doi ng your usual activities. At 7 o c' l ock in the evening you sit down to relax and notice you are fe el ing a little wheezy and breath less. What would you do? Prompt(i) Would you do anything else? (or anything after that?) (ii ) Would· you take any extra medication? (30 min) Over the next half hour the wheezing and breathlessness get worse and you find it a little difHcult to walk to the kitchen for a drink. What would you do? Prompt(i) Would you do anything else? (or anything else after that?) (ii) Would you take any extra rneclication?

Take oral steroids/action plan (only if not already taken ) Continue f) -agonist th erapy Maximum total

B. Rapid-Onset Attack (i) Take PEFH reading Take extra f) -agonist Hepeat measure of PEFH!check response Take increased dose of inhaled steroids (ii ) Take PEFR reading Action plan: Oral steroids and/or Call GP (home visit or urgent appointment) and/or Go immediately to ED Use f)-agonist Increase inhaled steroids (iii ) Call for ambulance Call GP (home visit) Go to ED, EMS, or GP (private transport) (Score 0 if drive th emselves) Take oral steroids/action plan (only if not already taken ) Use f)-agonist Ma.ximmn total

2

2

Scoring (Amended)

4 5

A. Slow-Onset Attack (i) Take peak expiratory flow rate (PEFR ) reading Use f)-agonist Hepeat measure of PEFH/monitor more closely Increase regular steroids (inhaled or oral) (Nil for oral ste roids ) (Nil for going to doctor) (ii ) Take PEFH reading Use ( + regu lar) f)-agonist Increase inhaled steroids Take oral steroids/action plan Consult general practitioner (GP), emergency medical services (EMS) or ED (within 24 to 48 h ) ( <24 hours if no steroids )

2 2 1 2

(iii ) Call ambulance - ED Call GP (to see at horne) Go to ED, EMS, GP, or Asthma Clinic (private transport) (Score 0 if drive the mselves)

4~}

90

2 2 3} 4 2

2 1 1 2 6}

2 1 7} 2 1

3 1 2.5

REFER ENCES

3 APPENDIX

25

*Braces indi cate mutually exclusive scores.

(1 h )

It has now been about 1 hand your breathing continues to get worse and by 8 o'clock you are so wheezy and breathless that you find it difficult to speak or get up from your chair. What would you do? Has anything like this ever happened to you? (Yes/No)

2

6 7 8

9 10 11 12 13

Bauman A. A meta-analys is of asthma patient education programmes [abstract]. Proceedings of the Thoracic Society of Australia and New Zealand Annual Scie ntific Meeting, Canberra, Australia; Mm-ch 30-April 2, 1992: 49 Garrett J, Fem\~ ck V, Taylor G, et al. Prospective controlled evaluation of the effect of a c:ommunity-based education centre: a multiracial working class neighbourhood. Thorax 1994; 49:976-83 Kolbe J,Garrett J, Vamos M, eta!. Influences on trends in asthma morbidity and mortality: the New Zealand experience. Chest 1994; l06(suppl): 211S-15S Sibbald B. Patie nt self-care in asthma. Thorax 1989; 44:97-101 Taylor GH , Rea HH, McNaughton S, et al. Atool for measming the asthma self-management c:ompetency of families. J Psychosam Res 1992; 35:483-91 Sears MH, Hea HH, Beaglehold R, etal. Asthma mortality in New Zealand: a two year national study. NZ M ed J 1985; 98:271-75 British Thoracic Association. Death from asthma in two regions of England. BMJ 1982; 285:125 1-55 Richards GN , Kolbe J, Fenwick J, et a!. Demographic characteristics of patients with severe l.ife-threatening asthma: c:omparison \vith asthma deaths. Thorax 1993; 48:1105-09 Woolcock A, Rubinfeld AH., Seale JP, et al. Asthma management plan. Med J Aust 1989; 151:650-53 Brewis G. Guidelines for the management of asthma in adults: I. Chronic asthma. BMJ 1990; 301:651-53 Brewis G. Guidelines for the manageme nt of asthma in adults: ll. Acute severe asthma. BMJ 1990; 301:797-800 Zigmond AS, Snaith HP. The hospital anxiety and depression scale. Acta Psychiatr Scand 1983; 67:361-70 O'Reilly P, Thomas HE. Hole of support networks in maintenance of improved cardiovascular health status. Soc Sci Med 1989; 28:249-60

Clinical Investigations