Abstract
Great emphasis is placed on educating asthmatics to use action plans to achieve better control of symptoms. The use of peak flow meters (PFM) has been recommended as an important part of self-management plans. We studied 92 (47 F) adult patients with asthma in a primary care setting to compare the effectiveness of action plans using either peak flow monitoring or symptoms to guide self-management. Each patient was instructed in the use of the action plan in the context of a 6-mo asthma education program taught by a nurse. Patients were already using inhaled corticosteroids or were newly prescribed corticosteroids by their family physician. Forty-four patients were randomized to the PFM group and 48 to the symptoms group. Spirometry, symptom scores, quality of life, medication use, and measures of health care utilization and morbidity (emergency department visits, hospitalizations, unscheduled doctor visits, and days lost from work or school) were recorded at baseline and throughout the study period. PC20 methacholine was measured at the first and at the final visits. There were significant improvements within groups for FEV1, symptoms score, PC20 methacholine, and quality of life, but no between-group differences. A significant shift from higher to lower daily use of β -agonists (p < 0.008 for both groups) and significant shifts to higher daily doses of inhaled steroids (p < 0.001) occurred in each group. Adherence to the self-management plans was only 65% in the PFM group and 52% in the symptoms group. Outcomes for health care utilization were similar except for fewer patients making unscheduled doctor visits within the PFM group. Our findings show that education, regular follow-up, and an action plan are effective in improving asthma control and quality of life, but the routine use of PFM to guide interventions is not the only way to accomplish these objectives.
Asthma self-management plans and education have been proposed as important components in the optimal management of asthma (1, 2). Self-management or action plans integrate changes in peak expiratory flow (PEF) measurements or symptoms with written directions to introduce or increase therapy. Recommendations to use peak flow meters (PFM) in the home management of asthma have been made in each of the most recent asthma consensus statements (3-8), but there have been only five randomized studies specifically addressing its usefulness (9-13). Two community-based studies (9, 10) found no difference between using PFM or monitoring symptoms, and another comparing PEF self-management with regular review found similar results (11). Hospital-based outpatient studies in Spain and in Finland did show significant differences for health care utilization (12, 13), spirometry (12), and quality of life (QOL) (13) in patients using a PFM and receiving education. However, the effect of PFM measurements on changes in self-management were not distinguished from the effects of regular follow-up and education in one study (12) and seemed to be correlated with the severity of symptom scores in the other (13).
Because of uncertainty about the benefits of using a PFM in self-management plans, we prospectively investigated the impact of self-management plans activated by PEF measurements or by symptoms. In addition to standard measures of morbidity and asthma control, we measured disease-specific QOL and bronchial hyperresponsiveness in adult patients with asthma recruited from a primary care clinic. We studied the null hypothesis of no difference between groups for outcome measures of morbidity (emergency department [ED] visits, hospitalizations, unscheduled doctor visits, days lost from school or work, courses of prednisone, respiratory tract infections) and QOL. The alternate hypothesis was that patients in the PFM self-management group would have significantly better outcomes.
This study was a prospective, randomized trial comparing two asthma self-management plans over a 6-mo period. All patients were given individualized asthma education. Patients were randomized to study groups using self-management plans activated by a decrease in PEF or an increase in symptoms (Table 1).
| Self Management Plan for PFM Group |
| 1. If PEF ⩾ 70% predicted, continue maintenance treatment: (1) bronchodilators as needed, (2) inhaled steroid twice daily. |
| 2. If PEF < 70% predicted: (1) double dose of inhaled steroid for number of days required to reach baseline PEF, (2) continue this increased dose for the same number of days needed to achieve baseline before, (3) returning to previous dose of maintenance treatment. |
| 3. If PEF falls to < 50% predicted: (1) start oral prednisone 40 mg daily after consulting with your family physician, (2) continue on this dose for the number of days required to achieve previous baseline for at least 1 wk, then (3) reduce oral prednisone by 5 mg daily until off. |
| 4. If PEF < 30% predicted: (1) contact family physician immediately or, if physician unavailable, (2) call ambulance (dial 911) or (3) go directly to hospital emergency department. |
| Self Management Plan for Symptoms Group |
| 1. When you feel normal continue maintenance treatment with: (1) bronchodilators as needed, (2) inhaled steroid twice daily. |
| 2. If you catch a cold or start to feel tight or awake at night with wheezing or have a persistent cough: (1) double the dose of inhaled steroid for the number of days it takes for you to return to normal, then reduce to maintenance dose of inhaled steroids after same number of days, (2) use bronchodilators two puffs every 4 h as needed. |
| 3. If the effect of your bronchodilators lasts only 2 h and you find doing your normal activities makes you short of breath (1) start oral prednisone 40 mg daily after consulting with your family physician, (2) continue to use this dose for at least 1 wk, or until symptoms have normalized, then reduce prednisone dose by 5 mg daily until off. |
| 4. If the effect of your bronchodilators lasts only 30 min or you have difficulty talking: (1) contact family physician immediately, or, if physician unavailable, (2) call ambulance (dial 911) or (3) go directly to hospital emergency department. |
Patients with a diagnosis of asthma defined by the American Thoracic Society were enrolled from a primary care clinic in Vancouver, British Columbia, Canada. The clinic is staffed by 21 family physicians and 14 specialists (no respirologists or allergists). Potential study patients were identified from the clinic computer database, and the clinic physicians were encouraged to refer patients meeting study criteria. We displayed a poster board and flyer advertisements in the clinic to encourage volunteers. All patients had written permission from their physician to participate. The protocol was approved by the University of British Columbia ethics committee, and written, informed consent was obtained from each patient.
Our intent was to enroll patients between 18 and 55 yr of age with moderate to moderately severe asthma to assess the effect of education and self-management plans. We defined asthma severity by including only patients with a baseline PC20 methacholine < 8 mg/ml and a daily requirement for inhaled corticosteroids to manage their asthma symptoms. Patients were either newly prescribed inhaled corticosteroids independently by their family physician or were currently using inhaled corticosteroids. Exclusion criteria included significant comorbid conditions that would impact on QOL measurements, current use of a PFM, inability to use a PFM, and inability to communicate in English.
Patients were randomized using a random number chart, after stratification for severity of airway responsiveness using values of PC20 methacholine < 2 mg/ml or ⩾ 2 mg/ml. We did not stratify for poor perception of symptoms because randomization should allocate these patients equally between the two groups. We estimated that 15% of patients would have impaired perception of asthma symptoms (14). On a pragmatic level, it is unlikely that primary care physicians would be able to identify all their patients with poor perception of dyspnea.
The clinic physicians were familiarized before the study with the self-management plans and the educational program. There was a general meeting with the principal investigator (J.M.F.), written information for each physician, and personal communication with the study nurse. We did not formally assess the clinic physicians understanding or previous use of self-management plans before the study.
The clinic provided an office for the full-time use of an asthma nurse educator. The nurse was experienced in the performance of spirometry, skin prick allergy testing, and methacholine inhalation challenges, and developed the education program in consultation with the physician coinvestigators.
The asthma nurse reviewed patients monthly for 6 mo after the initial visit (seven total visits). Patients who completed at least five visits were included in the final analysis. The initial visit included a history of asthma control and health care utilization for the previous 6 mo, spirometry, methacholine challenge testing, skin prick testing for common allergens, QOL, and instruction in the use of daily diary cards. The self-management plans and use of a PFM (when appropriate) were reviewed in detail after randomization. Monthly visits documented morbidity outcomes, reinforced and evaluated use of the self-management plan, and provided ongoing education. The self-management plans were similar to the plans used by Charlton and colleagues (9) modified to reflect current recommendations for the use of bronchodilators on an “as-needed” basis (6). Each subject's predicted peak flow measurement was used to determine the 70, 50, and 30% levels for institution of the management plan. Patients were asked to contact their physician if action points requiring prednisone were reached.
The primary study outcomes were: ED visits and/or hospitalization for asthma, unscheduled doctor visits, days absent from school or work, courses of prednisone, and respiratory tract infections. These prospectively documented events were also compared with events during the 6 mo before the study period. Exacerbations were defined by points in the self-management plans that required an increase in anti-inflammatory therapy. We identified nonadherence with the action plans as (1) increasing inhaled steroids outside the plan, (2) failure to increase inhaled steroids when indicated, (3) prescription for prednisone by the physician when not indicated by the plan, (4) failure of the physician to prescribe prednisone when indicated, and (5) failure to complete diaries.
We measured bronchial hyperresponsiveness at baseline and final visits by PC20 methacholine using the tidal breathing method described by Cockcroft and colleagues (15). Patients whose FEV1 fell 20% with saline inhalation were assigned a PC20 of 0.015 mg/ml, which is half the lowest concentration used (16). Spirometry and QOL were measured at each monthly visit, symptom scores (cough, nocturnal awakening, early morning chest tightness, and wheeze using a scale of 0 to 6), and medication usage were recorded daily in the patients' diaries.
We hypothesized that a study population recruited from a primary care setting rather than a specialist or hospital clinic would have milder asthma and fewer severe exacerbations requiring hospital care. We therefore decided a priori to incorporate a disease-specific QOL measure, the Asthma Quality of Life Questionnaire (17), in order evaluate the impact of asthma education and self-management plans on patients' overall lifestyles. The Asthma Quality of Life Questionnaire has 32 questions in four domains: activities, emotions, symptoms, and environmental triggers. The instrument has been validated, and it has been demonstrated to be responsive to change (18). The minimal clinically important difference is a change in score of 0.5 (on a scale of 7), a moderate change is 1.0, and a large change is 1.5 units (19).
A total of 150 patients were screened, and 117 patients were enrolled in the study. Twenty-five patients (16 randomized to symptoms and nine to PFM) were considered nonevaluable because of noncompliance with visits, diaries, or medications (n = 22) or they withdrew because of pregnancy, adverse effects (headaches attributed to inhaled steroid), and change of address. There was no significant difference in age, sex, and FEV1% predicted between groups for these 25 patients.
Education was provided to each patient who participated in the study. It was individualized, but the general format was based on disease characteristics, triggers for airway obstruction, and treatment objectives. The content of the educational program was developed from a review of the asthma education literature and from resource materials provided by Air Wise (2) and the Canadian Lung Association. There was an initial discussion of pathophysiology, the role and purpose of medications, recognition of worsening symptoms requiring implementation of the action plan, avoidance of triggers, and patient-doctor communication. Thirty minutes of education and discussion were allocated for each visit. Patients were encouraged to write in their diary specific questions they wanted to discuss at the monthly visit.
Group data are summarized as means and standard deviations (SD). Comparisons between groups for continuous variables (spirometry, PC20 methacholine, QOL, symptom scores, medication use) were analyzed by repeated measures analysis of variance. PC20 methacholine values were log-transformed for the purposes of analysis and expressed as geometric means. Baseline comparisons for steroid-naive and steroid-treated patients were analyzed by unpaired t tests. Dichotomous outcome measures were analyzed by chi-square or Fisher's exact tests. Shifts in the marginal distribution of proportions of patients using inhaled steroids and β-agonists before and after the study intervention were analyzed by Bhapkar's test of marginal homogeneity (21, 22). BMDP statistical software (version 7.0) was used for the analyses. A two-tailed p value of < 0.05 was considered to be statistically significant.
We evaluated 44 patients in the PFM group and 48 in the symptom group who attended the clinic for at least 4 mo after randomization (Table 2). Of the 92 total patients, there were 49 women (53%), and 80% of patients were either employed or students. No patients had a hospital admission for asthma in the 6 mo before randomization, although eight patients in the PFM and three in the symptoms group had attended the ED and were given treatment with inhaled bronchodilators and either oral or inhaled corticosteroids before discharge. There were no patients who had a previous history of mechanical ventilation or admission to an intensive care unit for severe asthma. Two patients had used and owned a PFM before entering the study but were not using it at the time of randomization. No patients had previously participated in a formal program of asthma education.
| PFM (n = 44) | Symptom (n = 48) | |||
|---|---|---|---|---|
| Sex, M/F | 22/22 | 21/27 | ||
| Age, yr* | 34.1 (10.5) | 34.1 (9.4) | ||
| Caucasian, % | 93 | 87 | ||
| Asthma duration, yr* | 17.9 (14.0) | 17.2 (13.5) | ||
| Atopic, % | 77 | 79 | ||
| Smoking, % | ||||
| Current | 22 | 23 | ||
| Ex-smoker | 43 | 40 | ||
| Never | 34 | 35 | ||
| Employment status, % | ||||
| Employed | 79 | 69 | ||
| Student | 9 | 8 | ||
| Unemployed | 9 | 19 | ||
| FEV1, L* | 2.84 (0.86) | 2.86 (0.88) | ||
| FEV1, % pred* | 78.1 (19.7) | 78.7 (18.9) | ||
| > 80% pred, n (%) | 24 (54.5) | 24 (50) | ||
| 60–80% pred, n (%) | 11 (25) | 18 (37.5) | ||
| < 60% pred, n (%) | 9 (20.5) | 6 (12.5) | ||
| FEV1/VC* | 0.70 (0.12) | 0.69 (0.19) | ||
| PC20, mg/ml† | 0.57 | 0.60 | ||
There was no difference between groups for duration of asthma, atopic status, smoking, or bronchial hyperresponsiveness (Table ). The baseline FEV1 %predicted (mean [SD]) was equivalent for the PFM (78.1% [19.7]) and symptoms (78.7% [18.9]) groups. Airflow obstruction (FEV1 < 80% predicted) was present at baseline in 45.5% of the PFM group and in 50% of the symptoms group. Perception of dyspnea, measured by Borg scores before the methacholine inhalation challenge, was similar between groups. At the PC20 methacholine, 20 patients, 11 in the PFM group (25%) and nine in the symptoms group (19%), had ⩽ 1 unit change on the Borg scale.
There was a significant increase in FEV1, PEF, and QOL and a decrease in symptom scores for both groups (all variables, p < 0.0001) after the first month of the study that was maintained throughout the study period (Table 3). There was no difference between groups for these variables (p > 0.39). The increase in FEV1 for all patients correlated inversely with change in symptoms (r = −0.40, p = 0.001) and positively with change in QOL (r = 0.34, p = 0.004).
| Visit 1 | Visit 2 | Visit 3 | Visit 4 | Visit 5 | Visit 6 | Visit 7 | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| FEV1, % pred | ||||||||||||||
| PFM | 78.1 | 85.6 | 87.2 | 88.8 | 88.5 | 87.5 | 83.0 | |||||||
| Symptoms | 78.7 | 87.3 | 87.8 | 88.3 | 88.7 | 88.4 | 86.1 | |||||||
| PEF, L/min | ||||||||||||||
| PFM | 368 | 435 | 432 | 445 | 435 | 431 | 406 | |||||||
| Symptoms | 370 | 429 | 415 | 420 | 424 | 429 | 410 | |||||||
| Quality of life* | ||||||||||||||
| PFM | 4.78 | 5.49 | 5.64 | 5.80 | 5.85 | 5.91 | 6.04 | |||||||
| Symptoms | 4.75 | 5.13 | 5.59 | 5.72 | 5.73 | 5.84 | 5.58 | |||||||
| Symptom score† | ||||||||||||||
| PFM | 8.2 | 4.9 | 4.8 | 4.7 | 4.1 | 4.2 | 3.2 | |||||||
| Symptoms | 9.1 | 5.8 | 4.0 | 4.5 | 4.5 | 4.6 | 5.2 | |||||||
Fourteen and 12 patients in the PEF and the symptom groups, respectively, were newly prescribed inhaled steroids. These steroid-naive patients had a higher baseline FEV1 (3.07 versus 2.76 L, p = 0.116) and had a greater increase in FEV1 (0.472 versus 0.194 L, p = 0.046) after 1 mo of inhaled steroid treatment (mean [SD], 616 [343] μg/d). Interestingly, patients already using inhaled steroids had a lower baseline FEV1 but fewer symptoms and higher QOL scores than did the steroid-naive patients at the first visit.
There was a greater than twofold increase in PC20 methacholine for both groups (p < 0.0001), from 0.57 to 1.23 mg/ml in the PFM group and from 0.60 to 2.64 mg/ml in the symptoms group. There were 12 and nine subjects in the PFM and symptoms groups, respectively, who had severe airway responsiveness (PC20 < 0.25 mg/ml) at baseline. There were only six patients (all in the PFM group) in the severe range at the final visit.
Quality of life scores showed a moderate increase for both groups, measured by change scores of greater than 1.0 from baseline. The improvement in QOL was greatest after 1 mo but continued throughout the first 5 mo of the study (Table ). The improvement in QOL was associated with an improvement in FEV1, PC20, and daily symptom scores, an increase in inhaled corticosteroid dose, and a decrease in inhaled β-agonist use for both groups.
There was a decrease in the use of inhaled β-agonists and an increase in the use of inhaled steroids (Figure 1). The lowest daily dose of β-agonists was at the fifth month, 239 μg/d (PFM) and 206 μg/d (symptoms). The mean daily dose of inhaled steroid increased steadily, with the largest change occurring at 1 mo, and a maximal mean daily dose after 5 mo, 986 μg/d (PFM) and 967 μg/d (symptoms). There was no significant difference between the groups, but we observed significant shifts to the left (from higher to lower daily doses) in the proportion of patients using β-agonists (Figure 2), (p = 0.007 and p = 0.002 for PFM and symptoms groups, respectively) and to the right (from lower to higher dose) for inhaled steroid (p < 0.001 for both groups) (Figure 3). The change in dose of inhaled steroids between first and last visits correlated with fewer symptoms (r = −0.30, p = 0.015) and an increase in FEV1 (r = 0.28, p = 0.025).
Fig. 1. Mean values of inhaled β2-agonists and corticosteroids from Visits 1 to 7 ( V1 to V7, monthly visits) for patients in the peak flow meter (PFM) and symptom (SYM) self-management groups. Mean doses of medication are expressed as micrograms per day. Beta2 = β2-agonists; ICS = inhaled corticosteroid.
[More] [Minimize]
Fig. 2. The proportion of patients in each group receiving daily doses of β2-agonists ranging from < 400, 400 to 999, 1,000 to 2,000 and > 2,000 μg/d. The daily doses were determined at Visit 1 (before randomization) and at the final visit. Mcg = micrograms.
[More] [Minimize]
Fig. 3. The proportion of patients in each group receiving daily doses of inhaled corticosteroids ranging from < 400, 400 to 999, 1,000 to 2,000, and > 2,000 μg/d. The daily doses were determined at Visit 1 (before randomization) and at the final visit. Mcg = micrograms.
[More] [Minimize]There were no group differences in health care utilization (Table 4). There were few exacerbations of asthma requiring hospital attendance, with only one hospitalization and 12 ED visits. This represents four fewer ED visits compared with the 6 mo before the study. There was an overall 49% decrease from the prestudy period for unscheduled doctor visits in the PFM group and a significant decrease in the number of PFM patients making visits (p = 0.038). Reported respiratory tract infections occurred with equal frequency in each group, but with an overall decrease of 21% from the 6 mo prior to the intervention.
| PFM Group(n = 44) | Symptom Group(n = 48) | |||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Before | During | Before | During | |||||||||||||
| (n) | (total events) | (n) | (total events) | (n) | (total events) | (n) | (total events) | |||||||||
| Hospitalizations | 0 | 0 | 0 | 1 | 1 | |||||||||||
| ED visits | 8 | 12 | 6 | 7 | 3 | 4 | 2 | 5 | ||||||||
| Days lost work or school | 9 | 29 | 9 | 37† | 10 | 32 | 8 | 38 | ||||||||
| Prednisone treatments | 5 | 5 | 3 | 4 | 3 | 3 | 6 | 6 | ||||||||
| Unscheduled doctor visits | 31 | 76 | 17 | 35‡ | 29 | 77 | 12 | 43 | ||||||||
| Respiratory tract infections | 39 | 109 | 35 | 94 | 43 | 110 | 36 | 80 | ||||||||
Each monthly diary was evaluated for adherence to the self-management plan. There was 65 and 52% adherence in the PFM and symptoms groups, respectively. Seventy-one percent of exacerbations (157 of 221) occurred in the symptoms action plan group of patients. Exacerbations identified by the action plan occurred in 56% of patients during the first month and decreased to 30% during the sixth month. Two or more exacerbations occurred in 16 and 38 patients in the PFM and symptoms groups, respectively. Change in therapy required by the management plans was appropriately implemented in 37.5% (24/64) of the PFM and 34.3% (54 of 157) of the symptoms exacerbations. There was no time effect observed in patients with multiple exacerbations. Prednisone was not ordered by physicians as required by the management plans in two PFM and in 10 symptoms patients.
This study of individual education and the use of self-management plans in outpatients with asthma shows a significant improvement within the PFM and symptom groups for measures of spirometry, airway responsiveness, symptoms, and QOL. However, there was no difference in primary outcome measures between the groups, which confirms the findings of Charlton and colleagues (9) in their study of asthmatic children and adults attending a general practice clinic in England and the Grampian Asthma Study (10) of 569 asthma outpatients in Scotland. The data from these studies (9, 10), our study, and those of Malo and colleagues (23) do not support the routine use of a PFM in all patients with asthma as a necessary part of their management plan. The use of either a symptom- or PEF-triggered action plan for patients attending a primary care clinic is likely to be equally effective when combined with individual asthma education.
The patients who participated in this study had moderate asthma manifested by their need for daily inhaled corticosteroids and as defined by consensus conference guidelines (5-8). Outcome measures of health care utilization and lost productivity from work or school are probably less sensitive for outpatient asthmatics participating in educational interventions or clinical trials of treatment. Therefore, our findings are more representative of the majority of asthmatics, and they explain the fewer event rates for health care utilization compared with educational and treatment interventions in patients with more severe asthma and frequent hospital admissions (24), recent admissions for asthma (25), and regular attendance at a hospital-based outpatient clinic (12). These studies also used a PFM and education as part of their interventions. There was significant decrease in health care utilization in each study, but the role of the PFM in contributing to these outcomes was not evaluated.
The use of a PFM has been recommended by consensus conferences for home-monitoring of PEF to maintain normal airway function (4, 5, 7), for patients with asthma receiving daily medications (8), and for patients requiring high dose inhaled steroids (⩾ 800 μg/d) (3). These recommendations are supported more by clinical impressions than by evidence, and they also have economic implications. If the routine use of a PFM is not a necessary or utilized part of a self-management plan, the provision of PFMs for all asthmatics in North America (5 to 10 million) would not be cost-effective. Evidence of improved outcomes or benefit from randomized trials comparing different treatment programs are necessary to demonstrate cost-effectiveness (26). Our findings do not support the routine use of PFMs in patients with few exacerbations requiring hospitalization or ED care.
The use of a PFM may not be the only way to achieve treatment objectives and reduce morbidity. Malo and colleagues (23) showed symptom recognition to be as effective as PEF monitoring in recognizing an exacerbation of asthma in patients with moderate to severe asthma. A study of exacerbations caused by reduction of inhaled steroid demonstrated that symptoms usually precede changes in inflammatory markers and measures of airflow obstruction (27). However, the sensitivity of symptoms to detect a change in asthma control depends on patient perception and the responsiveness of the instrument used to monitor symptoms in a diary or clinical trial. The symptom self-management plan used in our study may have been overly sensitive compared with the PEF action points, although the proportion of appropriate responses and outcomes were similar. We used a value of 70% of predicted PEF to initiate increased anti-inflammatory therapy, whereas more recent studies have used cut points of 80 to 85% (13, 28). Gibson and colleagues (29) have recommended using individualized PEF action points because they are more sensitive in identifying exacerbations and have fewer false positives than current recommendations for falls in PEF.
The use of both symptoms and PEF values has been recommended for self-management plans (30). Intuitively, this makes sense and solves the problem of poor perceivers being solely dependent on symptoms to identify deteriorations in asthma control. In an uncontrolled study of both symptoms and PEF values to guide self-management, 22 of 45 patients (48%) indicated a preference for having both instructions available when they experienced a “bad attack” of asthma (28). Data from Lahdensuo and colleagues (13) in Finland provide some interesting insights about the implementation of action plans. Only 32 of 56 patients (57%) had a fall in PEF of > 15% during a 1-yr period, and of these only 18 doubled their dose of inhaled steroid on each occasion. The implementation of the self-management instructions seemed to be associated with symptom severity (scale from 0 to 3), ranging from 29% when symptoms were absent to 100% when symptoms were scored as 3. These data demonstrate the futility of simply providing a PFM to poor symptom perceivers because self-management and better asthma control is unlikely to ensue without regular intensive follow-up and education.
The use of the PFM in asthma self-management is based on two premises: (1) PEF measurement is able to detect changes in asthma control and therefore reduce morbidity and mortality by the institution of increased therapy; (2) patients with asthma will be compliant with PEF monitoring and their action plan. Objective measurements of airflow obstruction are important and useful in assessing the severity of an exacerbation of asthma. The possibility that patients with severe asthma and poor symptom perceivers may benefit the most from home-monitoring of PFM needs to be studied prospectively. Garrett and colleagues (31) found only 16% of PFM owners measured their PEF before attending the ED. This may reflect a pattern of noncompliance or the difficulty of independent self-management for most patients. Our own experience in trying to educate patients with severe and near fatal asthma has been particularly disappointing, suggesting that patients who are most likely to benefit from the intervention are the least likely to comply (32).
An appropriate response to a deterioration in symptoms is important to effectively treat exacerbations in the majority of patients not monitoring PEF regularly. Scenarios designed to elicit responses for slow onset attacks of life-threatening asthma in well-educated asthmatics found that 54% would use a PFM and only 42% would increase their inhaled steroid dose (33). Overall, only a minority of our patients responded to a deterioration in asthma, as judged by their action plans, by changing therapy. However, if action plans are too sensitive there may be unwarranted increases in therapy or if insensitive there may be progression to a more severe attack.
Further research to clarify appropriate markers for increasing therapy from both a patient and physician perspective is needed. D'Souza and colleagues (28) found only 11 of 25 patients (44%) self-initiated a course of prednisone when indicated by a combined symptom and PEF action plan. In our study, patients were asked to contact their physician before taking oral corticosteroids. Implementation of prednisone therapy was indicated by the action plan but not prescribed by the physician in 10 of 16 patients (62.5%) in the symptoms group and two of five (40%) in the PFM group. This pattern of response suggests that symptoms alone may not be as convincing to the physician as objective measurements before prescribing oral corticosteroids for this patient population.
The greatest improvement in all measurements occurred in the first month of the study. This can be attributed to a combination of better compliance and response to the action plans by increasing inhaled steroid use in patients already using steroids and introducing therapy in the steroid-naive. We selected a 6-mo study period because it is a reasonable period of time to expect an impact from education and self-management and to minimize the inaccurate recall of events for a longer time period before the study intervention. Ignacio-Garcia (12) also observed a continued improvement in FEV1 in the patients using a PFM action plan but a subsequent fall in this parameter in the control group after the first month. There was no significant difference in medication use between these patient groups, suggesting that clinical improvements were achieved by better compliance through regular follow-up and monitoring. The early improvement observed in both studies is evidence that most patients will benefit from an initial, intensive intervention supplemented with ongoing reinforcement, particularly at the time of an exacerbation.
Compliance is a problem for patients asked to monitor PEF over a prolonged period of time. Malo and colleagues (23) found that compliance with recording both morning and evening PEF for 5 mo was only 55%. Another study of 3 mo of PEF monitoring using a computer chip in the PFM to record data (chronolog) found compliance decreased rapidly after 2 wk of monitoring (34). Overall, only 44% of expected values were stored and seven of 20 patients (45%) invented values recorded in their diaries. These patients were asked to make only serial PEF recordings for a 3-mo period before their next review and were not asked to follow an action plan (34). Although a potential weakness in our study was the lack of objective evaluation of PEF recordings, measurements were associated with an action plan, and patients brought their diary cards for monthly review, both strategies that may enhance compliance. A PFM chronolog may be especially useful to quantitatively assess compliance and the accuracy of PEF measurements in patients with poorly controlled asthma. We did not correlate the compliance with PFM measurements with appropriate responses to the self-management plans, but use of chronolog technology will aid in defining further the usefulness of PEF monitoring in this setting.
The Asthma Quality of Life Questionnaire was responsive to change in this group of patients with moderate asthma. There was a moderate, clinically important improvement in QOL associated with increased anti-inflammatory therapy and improvement in FEV1, PC20, and symptoms. Measuring health care utilization, lost productivity, and severity of disease by need for a course of prednisone is less likely to detect clinically important difference between interventions for patients with asthma treated in a primary care setting because these events are infrequent. Disease-specific QOL measurements incorporate symptom domains and therefore may be more sensitive than usual outcome measures of morbidity for patients with mild to moderate asthma.
To the best of our knowledge, this is one of the first studies of asthma self-management that also included measurements of airway hyperresponsiveness. There was a significant shift towards less airway hyperresponsiveness in both patient groups. Airway inflammation measured by sputum eosinophilia has been shown to correlate with airway hyperresponsiveness (35). Haahtela and colleagues (36) observed a significant decrease in airway hyperresponsiveness in patients with mild asthma treated with 1,200 μg/d of budesonide compared with regular terbutaline only after 6 wk of treatment. A double-blind randomized trial using 400 μg/d of budesonide for 12 mo found the greatest improvement of airway responsiveness at 3 mo, but continued improvement in some patients throughout the year (37). The improvement in airway hyperresponsiveness observed by ourselves and by others (36, 37) corresponds to changes in therapy, particularly increased therapy with inhaled steroids.
Our findings demonstrate the effectiveness of an asthma nurse working closely with physicians in a primary care setting. This is a model of shared care in which a specialist nurse educator is able to meet many patient needs and enhance the information available to the family physician. There was a significant decrease in the number of patients in the PFM group making unscheduled doctor visits and fewer total doctor visits for both groups. Nurses can deliver education less expensively than physicians can on an hourly fee basis, but frequent visits and telephone calls increase overall cost. Although visits to the nurse exceeded the total prestudy physician visits, issues of cost cannot be discussed without a formal economic evaluation. It would be important to compare important outcomes with a control group that could be evaluated without changing physicians' patterns of standard care.
We have shown that asthma education and an increase in anti-inflammatory therapy improved symptom scores, spirometry, bronchial hyperresponsiveness, and QOL in asthmatic patients attending a primary care clinic. The regular use of a PFM did not demonstrate any advantage over responding to symptoms in this management setting. Adherence to the self-management plans was uniformly poor and reflects the difficulties in implementing a process to achieve behavioral change for patients (38) and applying asthma guidelines by physicians (39). Notwithstanding the poor adherence to self-management action points, the positive findings of this study suggest that regular, accessible follow-up and increased therapy with inhaled steroids are the important features of successful asthma management. Education may be important in promoting compliance with prescribed therapy and reinforcing the need for adequate follow-up.
The writers thank the patients who willingly participated and Drs. Roy Oakey and Blake Wright and all the physicians and staff at the Seymour Clinic, Vancouver, for their assistance and support throughout this study. They also thank Dr. Michael Schulzer, Jason Peng, and Kukuh Noertjojo for statistical assistance.
Supported in part by a grant from Glaxo Wellcome Canada Inc.
| 1. | Clark N. M.Asthma self management education: research and implications for clinical practice. Chest95198911101113 |
| 2. | Beasley R., Cushley M., Holgate S. T.A self management plan in the treatment of adult asthma. Thorax441989200204 |
| 3. | Ernst P., FitzGerald J. M., Spier S.Canadian asthma consensus conference: summary of recommendations. Can. Respir. J.3199689100 |
| 4. | Woolcock A., Rubinfeld A. R., Seale J. P., Landau L. L., Antic R., Mitchell C., Rea H. H., Zimmerman P.Asthma management plan. Med. J. Aust.1511989650653 |
| 5. | British Thoracic Society, British Paediatric Association, Royal College of Physicians of London, The Kings Fund Centre, The National Asthma Campaign, Royal College of General Practitioners, General Practitioners in Asthma Group, British Association of Accident and Emergency Medicine, British Paediatric Respiratory GroupGuidelines on the management of asthma. Thorax481993S1S24 |
| 6. | Hargreave F. E., Dolovich J., Newhouse M.The assessment and treatment of asthma: a conference report. J. Allergy Clin. Immunol.85199010981111 |
| 7. | National Institutes of Health. 1991. Guidelines for the Diagnosis and Management of Asthma. National Asthma Education Program, Bethesda, MD. Publication No. 91-3042. |
| 8. | National Heart, Lung, and Blood Institute, National Institutes of HealthInternational consensus report on diagnosis and treatment of asthma. Eur. Respir. J.51992601641 |
| 9. | Charlton I., Charlton G., Broomfield J., Mullee M. A.Evaluation of peak flow and symptoms only self management plans for control of asthma in general practice. B.M.J.301199013551359 |
| 10. | Grampian Asthma Study of Integrated Care (GRASSIC)Effectiveness of self monitoring of peak flow in patients with asthma. B.M.J.3081994564567 |
| 11. | Jones K. P., Mullee M. A., Middleton M., Chapman E., Holgate S. T.the British Thoracic Society Research CommitteePeak flow based asthma self-management: a randomized controlled study in general practice. Thorax501995851857 |
| 12. | Ignacio-Garcia J. M., Gonzalez-Santos P.Asthma self-management education program by home monitoring of peak expiratory flow. Am. J. Respir. Crit. Care Med.1511995353359 |
| 13. | Lahdensuo A., Haahtela T., Herrala J., Kava T., Kiviranta K., Kuusisto P., Perämäki E., Poussa T., Saavelainen S., Svahn T.Randomized comparison of guided self management and traditional treatment of asthma over one year. B.M.J.3121996748752 |
| 14. | Rubinfeld A. R., Pain M. C. F.Perception of asthma. Lanceti1976882884 |
| 15. | Cockcroft D. W., Killian D. N., Mellon J. J. A., Hargreave F. E.Bronchial reactivity to inhaled histamine: a method and clinical survey. Clin. Allergy71977235243 |
| 16. | Brand P. L. P., Kerstjens H. A. M., Postma D. S., Sterk P. J., Quanjer P. H., Sluiter H. J., Dijkman J. H., van Herwaarden C. L., Hilvering C., Jansen H. M.et al. Long-term multicentre trial in chronic non-specific lung disease: methodology and baseline assessment in adult patients. Eur. Respir. J.519922131 |
| 17. | Juniper E. F., Guyatt G. H., Ferrie P. J., Griffith L. E.Measuring quality of life in asthma. Am. Rev. Respir. Dis.1471993832838 |
| 18. | Juniper E. F., Guyatt G. H., Epstein R. S., Ferrie P. J., Jaeschke R., Hiller T. K.Evaluation of health related quality of life in asthma: development of a questionnaire for use in clinical trials. Thorax4719927683 |
| 19. | Juniper E. F., Guyatt G. H., Willan A.Determining a minimal important change in a disease-specific quality of life questionnaire. J. Clin. Epidemiol.4719948187 |
| 20. | McNabb W. L., Wilson-Pessano S. R., Hughes G. W., Scamagas P.Self-management education of children: AIRWISE. Am. J. Public Health75198512191220 |
| 21. | Bhapkar V. P.A note on the equivalence of two criteria for hypotheses in categorical data. J. Am. Stat. Assoc.611966228235 |
| 22. | Stuart A.A test for homogeneity of the marginal distributions in a two-way classification. Biometrika421955412416 |
| 23. | Malo J. L., L'Archeveque J. L., Trudeau C., d'Aquino C., Cartier A.Should we monitor peak expiratory flow rates or record symptoms with a simple diary in the management of asthma? J. Allergy Clin. Immunol.911993702709 |
| 24. | Mayo P. H., Richman J., Harris H. W.Results of a program to reduce admissions for adult asthma. Ann. Intern. Med.1121990864871 |
| 25. | Yoon R., McKenzie D. K., Miles D. A., Bauman A.Characteristics of attenders and non-attenders at an asthma education programme. Thorax461991886890 |
| 26. | Krahn, M. 1994. Issues in the cost-effectiveness of asthma education. Chest 106(Suppl.):264S–269S. |
| 27. | Gibson P. G., Wong B. J., Hepperle M. J., Kline P. A., Girgis-Gabardo A., Guyatt G., Dolovich J., Denburg J. A., Ramsdale E. H., Hargreave F. E.A research method to induce and examine a mild exacerbation of asthma by withdrawal of corticosteroid. Clin. Exp. Allergy221992525532 |
| 28. | D'Souza W., Crane J., Burgess C., Te H., Karu, Fox C., Harper M., Robson B., Howden-Chapman P., Crossland L., Woodman K., Peane N., Pomare E., Beasley R.Community-based asthma care: trial of a “credit card” asthma self-management plan. Eur. Respir. J.7199412601265 |
| 29. | Gibson P. B., Wlodarczyk J., Hensley M. H., Murree-Allen K., Olson L. G., Saltos N.Using quality-control analysis of peak expiratory flow recordings to guide therapy for asthma. Ann. Intern. Med.1231995488492 |
| 30. | Fishwick D., Beasley R.Use of peak flow-based management plans by adult asthmatic patients. Eur. Respir. J.91996861865 |
| 31. | Garrett J., Mercer-Fenwick J., Taylor G., Mitchell E., Rea H.Peak expiratory flow meters (PEFMs): who uses them and how does education affect the pattern of utilization? Aust. N.Z. J. Med.241994521529 |
| 32. | Turner M. O., Crump S., Contrearas G. R., FitzGerald J. M.The challenge of delivering asthma education to a high risk group (abstract). Am. J. Respir. Crit. Care Med.1491994A251 |
| 33. | Kolbe J., Vamos M., James F., Elkind G., Garrett J.Assessment of practical knowledge of self-management of acute asthma. Chest10919968690 |
| 34. | Verschelden P., Cartier A., L'Archeveque J., Trudeau C., Malo J.-L.Compliance with and accuracy of daily self-assessment of peak expiratory flows (PEF) in asthmatic subjects over a three month period. Eur. Respir. J.91996880885 |
| 35. | Turner M. O., Hussack P., Sears M. R., Dolovich J., Hargreave F. E.Exacerbations of asthma without sputum eosinophilia. Thorax50199510571061 |
| 36. | Haahtela T., Jarvinen M., Kava T., Kiviranta K., Koskinen S., Lehtonen K.et al. Comparison of a β2-agonist, terbutaline, with an inhaled corticosteroid, budesonide, in newly detected asthma. N. Engl. J. Med.3251991388392 |
| 37. | Juniper E. F., Kline P. A., Vanzieleghem M. A., Ramsdale E. H., O'Byrne P. M., Hargreave F. E.Effect of a long-term treatment with an inhaled corticosteroid (budesonide) on airway hyperresponsiveness and clinical asthma in non-steroid-dependent asthmatics. Am. Rev. Respir. Dis.1421990832836 |
| 38. | Klingelhofer E. L., Gershwin M. E.Asthma self-management programs: premises, not promises. J. Asthma25198889101 |
| 39. | FitzGerald J. M., Spier S., Ernst P.Evidence based asthma guidelines. Chest110199613821383 |
