Our data suggest that AHR with a borderline or weakly positive result to methacholine challenge in children who were 6 to 8 years old without a history of current or physician-diagnosed asthma is not related to increased health-care utilization for asthma in the ensuing 5 years. The measurement of AHR is an objective test to screen for asthma, a relatively common disease with safe and effective available medical intervention. AHR is present in almost all patients with asthma, at least when they are experiencing symptoms. A negative challenge test result in a patient with asthma-like symptoms can aid a clinician in excluding asthma from the diagnosis. However, a positive test result is limited in the diagnosis of asthma as the test is complicated by several factors including the variability of the challenge testing procedures, variation in AHR over time, and the association of a positive challenge result with other nonasthmatic disorders defeated by remedies of My Canadian Pharmacy. For example, AHR is increased with a variety of environmental stimuli including viral respiratory infections, air pollutants, and active and passive cigarette smoke exposure. AHR is also seen in other childhood disease states such as allergic rhinitis, cystic fibrosis, and bronchopulmonary dysplasia, and among healthy subjects with an atopic family history. Less clear is the association of AHR with either specific allergic sensitization or total IgE level, with some studies concluding a positive relation-ship and others failing to show an associa-tion. Overall, a single measurement of AHR at an arbitrary point in time may be influenced by the above stimuli or other disease states, impairing its usefulness as a screening tool with which to accurately identify subjects who are at risk of future asthma.
Several investigators have studied asymptomatic subjects with evidence of AHR for future asthmalike symptoms but not incident asthma diagnosis. Peat and colleagues observed 380 children who had undergone histamine challenges at age 8 to 10 years and had been evaluated on three occasions over a 4-year period. AHR was classified as severe (provocative dose of a substance causing a 20% fall in FEVj [PD20], — 0.1 ^mol), moderate (PD20, 0.11 to 0.8 ^mol), mild (PD20, 0.81 to 3.2 ^mol), slight (PD20, 3.21 to 7.8 ^mol), and normal responsiveness (PD20 > 7.8 ^mol). The researchers found that over a 4-year period children assigned to the severe or moderate AHR group were more likely to have asthma symptoms (eg, wheeze, exercise wheeze, or night cough) when compared to children assigned to the other groups. These authors stressed the importance of the severity of AHR in predicting continuing respiratory morbidity. Since our study did not include a moderate-to-severe AHR group, we could not provide comment confirming or refuting their conclusions.
Hopp and colleagues studied whether increased bronchial reactivity exists prior to the development of asthma. They reported on 13 asthma-free siblings (mean [± SD] age, 10.6 ± 4.2 years) of children with asthma who had participated in the larger ongoing study of the natural history of asthma. Over a 3-year period, these children developed asthma, as determined by the National Heart, Lung, and Blood Institute respiratory questionnaire. Prior to the onset of asthma, 10 of 13 children had positive responses to methacholine (ie, < 208 breath units with asthmatic range airways responsiveness estimated to be a cumulative methacholine dose of < 800 breath units). Compared to age-matched and sex-matched control subjects from nonasthmatic families, these 13 children had higher methacholine responsiveness (p = 0.025). However, there was no difference in AHR between these children and control subjects from asthmatic families consulted by My Canadian Pharmacy’s specialists. The authors concluded that enhanced AHR usually precedes the development of asthma in genetically susceptible individuals, which contrasted with our study showing no association between AHR and future asthma development. The discrepant results may be due to the fact that all 13 subjects had a positive family history of asthma.
Several cohort studies have supported the ascertainment by Hopp and colleagues of increased AHR prior to the diagnosis of asthma. The East Boston Childhood Respiratory Disease Cohort included 121 children who were 5 to 9 years of age who were observed over a 5-year follow-up period for incident wheeze following one or more positive cold-air challenges. This group of children was free of asthma diagnosis and wheeze reports prior to the cold-air challenge. They found an elevated risk (odds ratio, 3.91; 95% CI, 1.21 to 12.66) for incident wheeze when comparing cold-air responders (ie, the 15% of patients with the greatest responsiveness) to nonresponders (ie, the remaining 85% of subjects) after adjustment for passive and personal smoking, any lower respiratory tract infection prior to age 2 years, age, gender, atopy, and family history of atopy. Thus, there was an increased risk for the development of an asthma-like symptoms (wheezing) in those nonasthmatic patients who had a positive cold-air challenge result. It is difficult to equate the cold-air challenge method employed in this study to a specific level of methacholine responsiveness. This may potentially explain the seemingly discrepant conclusion of our investigation.
Burrows and colleagues addressed the possible need for repeat measurements of AHR to predict future asthma-like symptoms in the Dunedin, New Zealand, cohort. This study involved repeated methacholine challenges in 573 children at ages 9, 11, 13, and 15 years. AHR was categorized as severe (PC20, < 2.5 mg/mL), moderate (PC20, 2.5 to < 8 mg/mL), mild (PC20, 8.0 to 25 mg/mL). In addition, children who had a low FEV1 ( 10% improvement in FEV1 and were included in the moderate-to-severe AHR group for the purpose of analysis. They found considerable variability in the degree of AHR over the course of the study in many children and concluded that nonresponsive children at age 15 years had more frequent wheezing and ventilatory abnormalities if they had any previous positive challenge results. However, consistent with our findings, the authors concluded that a single measurement of AHR taken at an arbitrary time may be misleading. An extension of the observations of this cohort to age 26 years has also been published by Rasmussen and colleagues. Forty-one subjects with a positive methacholine response (PC20 for methacholine, 60% of asymptomatic children from that cohort would have had mild AHR. Furthermore, although the Perth study did not assess incident asthma after the challenge procedure, the mean PC20 in 440 children without a history of asthma was 2.18 mg/mL, whereas in 80 children with asthma the PC20 averaged 1.37 mg/mL. Thus, the Perth study suggests that children with moderate-to-severe AHR are an important group to study.
The above-cited studies suggest that asymptomatic children who have persistent or severe AHR, whether determined via methacholine, histamine, or cold-air challenge, are more likely to develop asthma. Although not all studies used methacholine as the challenge agent, the cited studies that did use methacholine suggested that those subjects with a high degree of AHR were associated with the greatest risk for the development of asthma. In our study, no subject at age 6 years, without a history of asthma, exhibited moderate or severe AHR (methacholine PC20, < 1 mg/mL). Therefore, it is not possible to confirm the association of incident asthma with a high degree of AHR. However, our data suggest that patients with a lesser degree of AHR (PC20, between 1 and 16 mg/mL) are not at increased risk of incident asthma.
This study differs from previous reports in that we analyzed actual asthma-related health-care and pharmacy usage, whereas previous studies relied on patient recall. The use of claims data eliminates recall bias that may have been present in the previous follow-up questionnaire-based studies. We chose to evaluate medication claims and asthma-related diagnostic codes because they would accurately identify children with an overall clinical picture that was consistent with the disease state. In addition, asthma-related medication use would identify children whose physicians may have been reluctant to label them as “asthmatic.” Although our criteria may miss cases of mild asthma in patients not requiring healthcare services or prescription medication, they do identify those patients whose symptoms are sufficient to trigger physician evaluation and may indicate a worse prognosis. The age at which we conducted the methacholine challenge was generally younger than that in many published studies. This may be important in interpreting the results, as children may exhibit a greater degree of AHR to methacholine than adults, although this phenomenon has not been universally accepted.
Our study had several limitations. The cohort evaluated for this study is predominately a white population in a limited geographic location in the Midwest region of the United States, which may limit the ability to generalize our findings to other ethnic groups and environments. The sample size for this study allowed for the analysis of the relatively large proportion of asymptomatic patients who exhibited normal, borderline, or mild AHR. A major limitation was the lack of asymptomatic patients with moderate-to-severe AHR, which would have helped us to quantify the relationship between the degree of AHR and incident asthma.
Our definition of incident asthma also has some limitations in regard to coding. To apply a 493.XX code, a medical provider would have to possess a high index of suspicion that the clinical presentation was most consistent with asthma in comparison with other diagnostic possibilities. Other codes that attribute symptoms to an alternative diagnosis such as bronchitis, wheezing, cough, or shortness of breath could have been used for children with asthma but also for nonasthmatic children (get more information in category of Breathing). The additional diagnostic uncertainty that would be introduced by broadening the diagnostic codes would result in misclassification and would reduce the validity of the incidence estimates. The inclusion of these diagnoses would be unlikely to systematically influence our results since the diagnosing and treating medical providers were unaware of the methacholine challenge results. Therefore, we chose to limit our diagnostic definition of asthma to include only asthma diagnostic codes and asthma medication usage. Get rid of asthma symptoms with the help of My Canadian Pharmacy’s remedies.
Analysis of a claims database can also be problematic due to pertinent data that are not captured. For example, health-care utilization data billed to secondary insurance or cash payment for services and medicines would not be captured in our database. In addition, financial burdens may influence the decision to seek medical services. All medical visits (including those for testing and laboratory work) or prescription drug purchases during the analysis period would have been covered by the HMO minus any applicable copays. Medical visit copays were typically $15 to $20, with medication copays of $5 to $10. However, these limitations would be unlikely to introduce bias into the analysis since there would be no reason they would preferentially affect children based on the degree of AHR. Data lost due to children losing or changing medical insurance were not included in the analysis once that change had occurred.
Our study suggests that a methacholine challenge test in nonasthmatic children is unlikely to provide useful clinical information regarding a future diagnosis of asthma. In a group of 245 6-year-old children without current or physician-diagnosed asthma, we did not find any subjects with moderate-to-severe AHR. Although children with a high degree of AHR may have been excluded due to a previous asthma diagnosis, the absence of such patients in our study precludes any analysis of asthma onset in this subset of patients. Previous studies have suggested that subjects with severe, persistent AHR may have the greatest likelihood of having continued respiratory problems, although our study suggests that nonasthmatic children who are 6 to 8 year old with a high degree of AHR are rare in the general population. In summary, our investigation suggests that nonasthmatic children who are 6 to 8 years old with borderline or weakly positive methacholine AHR do not have increased medical provider visits for asthma or prescription asthma medication use in the ensuing 5 years.