Do written action plans improve patient outcomes in asthma? An evidence-based analysis
The Journal of Family Practice. 2002 October;51(10):842-848
October 1, 2002|Family Medicine
Author and Disclosure Information
- OBJECTIVE: Current guidelines recommend use of written action plans and peak flow monitoring as key components of asthma care. Our study assesses whether written action plans, with or without peak flow monitoring, have an independent effect on outcomes when used as a component of asthma self-management.
- STUDY DESIGN: This was a systematic review of published studies. Two independent reviewers followed a prospective protocol for study selection and data abstraction. Outcome data were synthesized qualitatively; they were not appropriate for quantitative meta-analysis. Our comprehensive literature search used MEDLINE, Embase, the Cochrane Library, and a hand search of recent bibliographies. The search was limited to full-length, peer-reviewed articles with abstracts in English. The studies were randomized controlled trials that compared the outcomes of an asthma self-management intervention with and without the use a written action plan. The primary outcomes of interest are utilization measures, such as hospitalizations and ER visits. Other outcomes of interest include measures of symptom control and lung function
- POPULATION: There were 1501 evaluable patients with asthma; 1410 adults and 91 children.
- OUTCOMES MEASURED: We measured the frequency of waiting and examination room companions, the reasons for accompaniment, the influence on the encounter, and the overall helpfulness of the companion as assessed by patients and companions. We also determined the physician’s assessment of the companion’s influence, helpfulness, and behavior during the encounter.
- RESULTS: Nine randomized controlled trials enrolling a total of 1501 patients met selection criteria. The majority of comparisons in these studies do not demonstrate improved outcomes associated with a written action plan. There are notable methodologic limitations: studies reporting negative findings lack sufficient power, and studies reporting positive findings demonstrate systematic bias.
- CONCLUSIONS: Although written action plans are widely used, there is insufficient evidence to determine whether their use, with or without peak flow monitoring, improves outcomes.
TABLE 1
Study characteristics
| Study | Patient popultation | Study Arms | Intervention components | Outcomes reported | Asthma quality indicators met |
|---|---|---|---|---|---|
| Optimal medical management vs. optimal medical management + PFM action plan | |||||
| Jones 199514 | Inclusions: patients using ICS | Usual care | SxD, FU | Ut, LF, Sx | Pow, Med |
| Exclusions: patients on oral steroids or using peak flow meters at home | PFM action plan | AP, PF, SxD, FU | |||
| Mean age: 29.5 years | |||||
| Severity level: Mild–moderate | |||||
| Drummond 1994 (GRASSIC)15 | Inclusion: FEV1 reversibility 20% or greater | Usual care | FU | Ut, LF, Med Ex | Pow, Rev |
| Exclusions: patients who already owned a PFM | PFM action plan | AP, PF, FU | |||
| Mean age: 50.8 years | |||||
| Severity level: Mild–severe | |||||
| Ayres 199516 | Inclusions: maximum PEF variability, 0.15%; minimum nights/week with symptoms, 3; minimum use of ICS or sodium cromoglycate, 3 months | Usual care | SxD, FU | LF, Sx, Ex | Pow, Med |
| Mean age: 45 years | PFM action plan | AP, PF, SxD, FU | |||
| Severity level: Moderate–severe | |||||
| Cowie 199713 | Inclusions: treatment for an exacerbation of asthma in an ER asthma clinic; history of receiving urgent treatment for asthma in the previous 12 months | Usual care | Ed, SxD, FU | Ut, PF, Med, Ex | None |
| Mean age: 37.8 years | PFM action plan | AP, PF, Ed, SxD, FU | |||
| Severity level: Mild–severe | |||||
| Cote 199717 | Inclusions: FEV1postbronchodilator 85-100 % of predicted; PEF, at minimum, 85 % of predicted; minimum PEF variability, 0%; Methacholine | Usual care | Ed | Ut, LF, Med | Exc, Rev, Com |
| Exclusions: patients having previously taken an asthma educational program | PFM action plan | Ed, Cn, AP, PF | |||
| Mean age: 36.5 years Severity level: Mild | |||||
| Usual care + PFM use alone vs. usual care + PFM action plan | |||||
| Ignacio-Garcia 199518 | Inclusions: patients from outpatient asthma clinic with asthma for 2 years | Usual care + PFM | PF, SxD, FU | Ut, LF, Med | None |
| Mean age: 41.9 years | Usual care + PFM action plan | PF, AP, Ed, SxD, FU | |||
| Severity level: Mild–severe | |||||
| Charlton 199419 | Inclusion: patients with inpatient or outpatient visit for asthma | Usual care + PFM | PF, Ed, SxD, FU | Ut, Sx, Med, Ex | None |
| Mean age: 6.5 years | Usual care + PFM action plan | PF, AP, Ed, SxD, FU | |||
| Severity level: Mild–moderate | |||||
| PFM action plan vs. Symptom action plan | |||||
| Turner 199820 | Inclusions: Maximum methacholine PC20, 7.9; using ICS | Symptom action plan | AP, Ed, SxD, Cn BM, EM | Ut, LF, Sx, Med | Exc, Com |
| Exclusions: previous PFM use; significant comorbid conditions | PFM action plan | PF, AP, Ed, SxD, Cn BM, EM | |||
| Mean age: 34.1 years | |||||
| Severity level: Mild–severe | |||||
| Charlton 199021 | Inclusions: patients on repeat prescribing register | Symptom action plan | AP, Ed, FU | Ut, Med | None |
| Mean age: NR | PFM action plan | PF, AP, Ed, FU, Cn | Ut, PF, Med, Ex | None | |
| Severity level: Mild–severe (?) | |||||
| Cowie 199716 | Inclusions: treatment for an exacerbation of asthma in an ER, or asthma clinic; history of receiving urgent treatment for asthma in the previous 12 months | Symptom action plan | AP, Ed, SxD, FU | ||
| PFM action plan | AP, PF, Ed, SxD, FU | ||||
| Cote 199717 | Inclusions: FEV1postbronchodilator, 85-100 % of predicted; PEF, at minimum, 85 % of predicted; minimum PEF variability, 0%; Methacholine | Symptom action plan | Ed, AP | Ut, LF, Med | Exc, Rev, Com |
| Exclusions: previous enrollment in an asthma educational program | PFM action plan | Ed, Cn, AP, PF | |||
| Eligibility criteria: ICS = inhaled corticosteroid; FEV1 = forced expiratory volume in 1 second; PEF = peak expiratory flow; PFM = peak flow meter; ER = emergency room; PC20 = 20% fall in FEV1 Intervention components: PF = Peak flow meter; AP = Written Action Plan; Ed = Education; SxD = Symptom diary; FU = Follow-up visits; Cn = Counseling; BM = Behavior modification; EM = Environmental modification | |||||
| Outcomes: Ut = Utilization measures; LF= Lung function measurements; Sx = Symptom=based measurements; Med = Medication use; Ex = Exacerbations of asth ma Asthma Quality Indicators: Exc = Accounted for excluded patients; Pow = Reported power calculations; Rev = Established reversibility of airway obstruction; Med = Controlled for other medication use; Com = Reported compliance; Sea = Addressed seasonality. | |||||
TABLE 2
Power calculations for hospitalizations per patient per year
| Assumed control mean | Possible treatment mean | % decrease | N needed per study arm |
|---|---|---|---|
| 0.10 | 0.075 | 25 | 3077 |
| 0.10 | 0.05 | 50 | 770 |
| 0.10 | 0.025 | 75 | 342 |
| 0.20 | 0.015 | 25 | 770 |
| 0.20 | 0.10 | 50 | 193 |
| 0.20 | 0.05 | 75 | 86 |
| 0.30 | 0.225 | 25 | 342 |
| 0.30 | 0.15 | 50 | 86 |
| 0.30 | 0.075 | 75 | 38 |
| Studies were identified that contained baseline rates on hospitalizations/patient/year, or information that allowed calculation of this parameter (Drummond, Abdalla, Beattie et al., 1994; Cote, Cartier, Robichaud et al. 1997; Cowie, Revitt, Underwood et al., 1997; Ignacio-Garcia and Gonzalez-Santos, 1995). Baseline rates of hospitalization varied in these studies from 0.04-0.29/patient/year. Standard deviations for this outcome were available only in two studies; Cote, Cartier, Robichaud et al. (1997) reported an SD of 0.30 for this variable, and an SD of 0.35 was calculated from the confidence intervals reported in GRASSIC (Drummond, Abdalla, Beattie et al., 1994). For the calculations, the more conservative 0.35 estimate for SD was used. | |||
| Number of patients per study arm were estimated for 80 percent power at the 5 percent significance level using control arm means of 0.10, 0.20, and 0.30 hospitalizations/patient/year. The expected reduction in this variable was tested along a spectrum from 25-75 percent. | |||
Written action plan versus no written action plan
All 5 studies used a peak flow meter based written action plan. All reported utilization outcomes, but the types and units of measurement were not consistent across studies (Table 2). Additionally, 4 studies reported on symptoms,13-16 and 3 reported lung function outcomes.13-15
