A Gap Between Asthma Guidelines and Management for Adolescents and Young Adults

18 years ( P [ .02). At least 1 dispensation of any inhaled corticosteroid before age 18 years was found for 73% (107 of 147), compared with 50% (74 of 147) after age 18 years. The mean number of dispensed any inhaled corticosteroid was 3.1 before 18 years and 2.1 after 18 years ( P < .01). Only 3% (5 of 147) had a regular dispensation of any inhaled corticosteroid once a year during the 8-year period. CONCLUSIONS


INTRODUCTION
Most patients with asthma have mild or moderate disease and can be managed in primary care; with currently available medications, most can be treated effectively. 1,2The long-term goal of asthma treatment is to achieve control of symptoms and maintain normal activity levels. 3It is therefore important to regularly monitor symptom control, risk factors, and response to treatment through follow-up visits performed by an appropriate suitable health care provider at an appropriate level of care. 3wedish recommendations on asthma management are similar to the Global Initiative for Asthma (GINA) guidelines.4][5] At around 18 years old, management involves a transition from pediatric to adult health care. 6Our recent qualitative study showed that young adults with severe asthma felt left out of the system during the transition from pediatric to adult health care. 7oreover, they experienced fewer or no follow-up visits in adult health care.Thus, it is relevant to investigate whether the transition to adult health care influences asthma-related health care consumption in a wider group.It is also important to explore how the process of transition affects pharmacological dispensation, because nonadherence to therapy is particularly common during this period. 8,9Medication adherence is the cornerstone for improving the patient's health-related quality of life, and an improved adherence can lead to decreased asthma morbidity and mortality. 10he aim of this study was therefore to investigate asthmarelated health care consumption and pharmacological dispensation during the transition from pediatric to adult health care.A longitudinal approach was used to follow different asthma phenotypes during the entire transition process.

Study design
During the period 1994 to 1996, parents to all newborns living in predefined areas of Stockholm, the capital of Sweden, including inner-city, urban, and suburban districts, were asked to participate in the longitudinal population-based birth cohort BAMSE (Barn/ Child, Allergy, Milieu, Stockholm, Epidemiology). 11,12The ongoing birth cohort includes 4089 participants, who have been followed since birth with repeated follow-ups.At age 2 months, a baseline questionnaire was answered.When the participants were approximately aged 1, 2, 4, 8, 12, and 16 years, parents completed follow-up questionnaires to collect information about symptoms related to asthma and other allergic diseases, lifestyle factors, and treatment of asthma.At 12, 16, and 24 years, participants were also asked to complete a questionnaire themselves.In addition to the questionnaires, the participants were invited to undergo clinical examinations, including blood sampling and lung function measurement, at approximately ages 4, 8, 16, and 24 years.
The study population consisted of 1808 participants who responded to the questionnaires and lived in the Stockholm region at both the 16-and 24-year follow-ups (Figure 1).The mean age was 16.5 years at the 16-year follow-up and 22.4 years at the 24-year follow-up.For analyses related to the clinical examinations, participants with a valid spirometry measure were included.

Lung function exposure assessment
At the clinical examinations carried out at approximately ages 16 and 24 years, lung function was measured through spirometry using a Jaeger MasterScreen-IOS system (Carefusion Technologies, San Diego, Calif).All subjects performed repeated maximal expiratory flow volume measurements. 13The highest values of forced vital capacity (FVC) and FEV 1 were extracted and used for analysis in accordance with the European Respiratory Society and American Thoracic Society criteria. 14FEV 1 /FVC data were converted to z scores on the basis of The Global Lung Function Initiative [GLI]reference values.†The FEV 1 /FVC ratio z score below the lower limit of normal, defined as the lower fifth percentile in the never-asthmatic population.
zDispensed at least 800 mg budesonide or equivalent, 500 mg fluticasone, or fixed combinations of ICSs and LABAs.
xAsthma with high daily doses of ICS plus dispensed LABA and/or leukotriene receptor antagonist at least once in the 18 mo before the 16-y follow-up to prevent the asthma from becoming or remaining uncontrolled, defined as at least 1 of the following 4 alternatives on the basis of data from the 16-y follow-up: (1) Uncontrolled asthma based on modified GINA definition 3,16 ; (2) Taken cortisone tablets dissolved in water for asthma or respiratory symptoms 3 d in a row; (3) Sought acute medical care because of respiratory symptoms; and (4) Impaired lung function, FEV 1 below 80% of predicted.jjP values obtained using tests of proportions and Wilcoxon signed-rank test indicated differences between 4 y before and after age 18 y for visits and level of care.

Asthma phenotypes-Exposure definitions
Exposure definitions were based on questionnaire and clinical data from the 16-and 24-year follow-ups.Current asthma was defined as fulfilling at least 2 of the following 3 criteria 16 : Symptoms of wheeze and/or breathing difficulties in the last 12 months, ever doctor's diagnosis of asthma, and/or asthma medication occasionally or regularly in the last 12 months.Persistent asthma was defined as fulfilling the definition of current asthma at both the 16-and 24-year follow-ups.
Allergic asthma was defined as a combination of asthma and IgE sensitization to inhalant allergens (cat, dog, horse, and/or house-dust mite, timothy grass, birch, mugwort, and/or mold). 17evere asthma was based on European Respiratory Society and American Thoracic Society and GINA guidelines, 1,3,16 defined as asthma with high daily doses of inhaled corticosteroids (ICSs).Information on pharmacy-dispensed asthma medication within the 18 months before each follow-up was obtained by linkage to the Swedish Prescribed Drug Register. 18High doses were identified when a participant was dispensed at least 800 mg budesonide or equivalent, greater than or equal to 500 mg fluticasone, or fixed combinations of ICSs and long-acting b 2 -agonists (LABAs), and dispensed LABAs and/or leukotriene receptor antagonists at least once to prevent the asthma from becoming or remaining uncontrolled despite therapy.Uncontrolled asthma was defined as at least 1 of the following 4 alternatives on the basis of data from the 16-or 24-year follow-ups: (1) Uncontrolled asthma based on modified GINA definition 3,16 ; (2) Taken cortisone tablets dissolved in water for asthma or respiratory symptoms 3 days or more in a row; (3) Sought acute medical care because of respiratory symptoms; and (4) FEV 1 below 80% of predicted.
Airflow obstruction was defined as the FEV 1 /FVC ratio z score below the lower limit of normal, defined as the lower fifth percentile in the never-asthmatic population.

Data sources of outcome assessment
A flowchart is shown in Figure 1 and presents how questionnaire and clinical data from the 16-and 24-year follow-ups were linked to mandatory Swedish health registries between 2008 and 2018: In total, an 8-year period (based on individual age, participants born between 1994 and 1996)-4 years before and 4 years after age 18 years, respectively.This period is hereafter denoted "before and after age 18 years."A timeline with study periods and data collection is presented in Figure E1 in this article's Online Repository at www.jaci-inpractice.org.Data on asthma-related health care consumption were obtained from Stockholm Regional Healthcare Data Warehouse Vårdanalysdatabasen (VAL). 19The VAL database includes complete data on all health care consultations in primary and specialist care, all hospitalizations and medical procedures, and diagnoses based on the International Classification of Diseases, Tenth Revision (ICD-10). 20or each health care consultation, a maximum of 15 diagnoses based on the ICD-10 can be registered.One diagnosis is assigned as the main condition, whereas others are secondary. 21We identified participants with physician-diagnosed asthma, ICD-10 codes J45 and/or J46, as main or secondary diagnosis.With data linked to the personal identity number, it is possible to follow each individual over time. 22nformation on dispensed asthma medications was obtained by linkage to the national Swedish Prescribed Drug Register using personal identity numbers. 18,22,23The medications included were the following, classified in accordance with the Anatomical Therapeutic Chemical Classification System 24 : short-acting b 2 -agonists , ICSs (R03BA), fixed combinations of ICSs and LABAs (R03AK), and leukotriene receptor antagonists (R03DC).

Covariates
Information on covariates was obtained from the baseline questionnaire (sex, mother's age at birth, parent born outside Sweden, parental allergic disease, parental education, and parental smoking).

Statistical analysis
Tests of proportions (categorical variables) and Wilcoxon signedrank test (continuous variables) were used to study differences between groups (background characteristics, consultations, levels of care, medical procedures, and dispensed asthma medications).Twosample Wilcoxon rank-sum test was used to study sex differences.
McNemar test was used to evaluate differences in asthma control over time.P values of less than .05were considered statistically significant.
The association between having had a consultation after age 18 years and selected asthma phenotypes (allergic asthma, asthma and airflow obstruction, asthma with high daily doses of ICS, and severe asthma) was analyzed using a logistic regression model.Potential confounders were selected a priori from the previous literature: sex and socioeconomic status and expressed as odds ratios (ORs) with 95% CI.
For the main analyses, we used ICD-10 codes J45 and/or J46.Sensitivity analysis with all ICD-10 codes J e "Diseases of the respiratory system" was performed to assess potential underreporting.
All analyses were performed with the STATA statistical software (release 14.2; College Station, Texas).
The study was approved by the Regional Ethical Review Board in Stockholm, Sweden.All participants and parents provided informed consent to participate in the study.

Health care consumption
Asthma at the 16-year follow-up.At the 16-year follow- up, 14% (n ¼ 253) of the adolescents fulfilled the study definition of current asthma (Table I).Of these, 62% (n ¼ 157) had allergic asthma, 7% (n ¼ 18) had airflow obstruction (FEV 1 / FVC z score < lower limit of normal), 24% (n ¼ 60) were dispensed high daily doses of ICSs or fixed combinations of ICSs and LABAs, and 5% (n ¼ 12) fulfilled the definition of severe asthma (Table II).
In the 4-year period before their 18th birthday, 32% (82 of 253) of the adolescents had at least 1 asthma-related consultation, compared with 27% (68 of 253) in the following 4-year period (Table I).The mean number of consultations decreased from 1.2 before 18 age years to 0.6 after age 18 years (P < .01).This relationship was seen for all asthma phenotypes (Table II).In the sensitivity analyses including consultations for all diagnoses of diseases of the respiratory system, the mean number of consultations before and after 18 age years was 3.0 versus 2.0 (P < .01)(see Table E2 in this article's Online Repository at www.jaci-inpractice.org).
Asthma at the 24-year follow-up.At the 24-year followup, 14% (n ¼ 248) of the young adults fulfilled the study  †A combination of asthma and IgE sensitization to inhalant allergens (cat, dog, horse, and/or house-dust mite, timothy grass, birch, mugwort, and/or mold).
zThe FEV 1 /FVC ratio z score below the lower limit of normal, defined as the lower fifth percentile in the never-asthmatic population.
xDispensed at least 800 mg budesonide or equivalent, 500 mg fluticasone, or fixed combinations of ICSs and LABAs.
jjAsthma with high daily doses of ICSs plus dispensed LABAs and/or leukotriene receptor antagonists at least once in the 18 mo before the 16-y follow-up to prevent the asthma from becoming or remaining uncontrolled, defined as at least 1 of the following 4 alternatives on the basis of data from the 16-y follow-up: (1) Uncontrolled asthma based on modified GINA definition  †The FEV 1 /FVC ratio z score below the lower limit of normal, defined as the lower fifth percentile in the never-asthmatic population.
zDispensed at least 800 mg budesonide or equivalent, 500 mg fluticasone, or fixed combinations of ICSs and LABAs.
xAsthma with high daily doses of ICSs plus dispensed LABAs and/or LTRAs at least once in the 18 mo before the 16-y follow-up to prevent the asthma from becoming or remaining uncontrolled, defined as at least 1 of the following 4 alternatives on the basis of data from the 16-y follow-up: (1) Uncontrolled asthma based on modified GINA definition 3,16 ; (2) Taken cortisone tablets dissolved in water for asthma or respiratory symptoms 3 d in a row; (3) Sought acute medical care because of respiratory symptoms; and (4) Impaired lung function, FEV 1 below 80% of predicted.jjP values obtained using Wilcoxon signed-rank test indicated differences between 4 y before and after age 18 y for respective dispensed asthma medicines.
Persistent asthma.Eight percent fulfilled the criteria for persistent asthma (Table I).Before age 18 years, 39% (58 of 147) of these young adults had 1 or more asthma-related consultation, similar to 37% (55 of 147) after age 18 years.Only 2% (3 of 147) had yearly consultations during the entire study period of 8 years.The mean number of consultations decreased from 1.6 before age 18 years to 1.0 after age 18 years (P ¼ .02).
Figure 2 shows the number of consultations before and after age 18 years, respectively, divided by level of care.The most common combination was having no consultation either before or after age 18 years, and the second most common was attending specialist care before age 18 years but having no consultation after age 18 years.After age 18 years, the mean number of consultations in primary care increased significantly and the number in specialist care decreased (Table I).The sensitivity analyses showed similar results (see Table E2).
Uncontrolled asthma was found among 57% (80 of 147) at the 16-year follow-up, and among 72% (103 of 147) at the 24year follow-up (P < .01).There were 2 registered emergency room visits before age 18 years, and 4 after age 18 years.During the study period of 8 years, there was 1 registered hospitalization with asthma as the main diagnosis, whereas 9 of the young adults had hospitalizations where asthma was a secondary diagnosis (8 on 1 occasion and 1 on 2 occasions).
Of all registered consultations during the study period, indirect contacts, for instance, by telephone and mail, accounted for a total of 4% (17 of 382).
Factors associated with the odds for respective asthma phenotypes (allergic asthma, asthma and airflow obstruction, asthma with high daily doses of ICSs, and severe asthma) and having a consultation after age 18 years among those with persistent asthma showed that asthma with high daily doses of ICSs was associated with increased odds of having a consultation (OR adj ¼ 2.6; 95% CI, 1.3-5.6)(Table III).Increased odds were also seen for severe asthma (OR crude ¼ 3.9; 95% CI, 1.0-16.0),but no significant association was seen in the adjusted model (OR adj ¼ 3.9; 95% CI, 0.9-16.1).No association was seen for asthma and airflow obstruction.
The mean number of registered spirometry tests in VAL decreased significantly, from 0.27 before age 18 years to 0.16 after age 18 years (P < .01).One or more spirometry was registered among 27% (40 of 147) of the young adults before 18 age years, and among 16% (24 of 147) after age 18 years.
Figure 3 shows the prevalence of current asthma, and the mean number of yearly consultations in relation to sex.Males had a higher mean number of consultations before age 18 years (males: 2.1, females: 1.2, P ¼ .25),but after age 18 years, no difference was seen (males: 0.9, females: 1.1, P ¼ .90).
zP values obtained from Wilcoxon signed-rank test indicated differences between 4 y before and after age 18 y for respective dispensed asthma medicines.
age 18 years for all asthma phenotypes, except severe asthma (Table IV).Among those with persistent asthma, at least 1 dispensation of SABA before age 18 years was found for 70% (103 of 147) compared with 50% (73 of 147) after age 18 years.The average number was 2.8 before age 18 years, and 2.1 after age 18 years (P < .01)(Table V).Only 3% (4 of 147) had a regular dispensation of SABA once a year during the entire study period of 8 years.At least 1 dispensation of any ICS before age 18 years was found for 73% (107 of 147), compared with 50% (74 of 147) after age 18 years.The mean number of dispensed any ICS was 3.1 before age 18 years and 2.1 after age 18 years (P < .01)(Table V).Only 3% (5 of 147) had a regular dispensation of any ICS once a year during the entire 8-year period.

DISCUSSION
This longitudinal population-based birth cohort, investigating different asthma phenotypes in relation to asthma-related health care consumption and pharmacological dispensation during the transition from pediatric to adult health care, showed that there is a gap between asthma guidelines and actual management.Almost two-thirds of the young adults with persistent asthma had not had any follow-up visit after age 18 years.For all asthma phenotypes, health care consultations were fewer than recommended in guidelines, and their frequency decreased after the transition.The dispensations of asthma medications decreased after the transition, even for the participants with severe asthma.For all asthma phenotypes, almost no one had dispensed regular asthma medications during the 8-year period.
Few previous studies have addressed the transition from pediatric to adult health care for patients with asthma.However, a recent French observational study characterized changes in asthma care in adult patients with persistent asthma, and found that the number of visits per year to specialist care increased with time during a 10-year period, whereas the number of visits to primary care decreased. 25In the present study, the opposite was seen regarding level of care.The result was expected, given that Swedish primary care is responsible for providing basic medical treatment. 26However, a large proportion of the participants attending specialist care before age 18 years had no consultation either in primary or in specialist care after age 18 years.Patients who are managed well in primary care can remain with their primary care physician, ensuring continuity, but most adolescents with asthma requiring a tertiary level of care would be expected to need specialist care as adults. 8One of few published studies examining the impact of randomized referral to either primary or specialist care, and risk factors for deterioration during the transition from pediatric to adult health care, showed that mild/moderate asthma was managed equally effectively regardless of level of care. 274][5] A recent US study assessed clinician-reported adherence to asthma guideline recommendations, and found that agreement with and adherence to guidelines was higher for specialty physicians than for primary care physicians, and overall low adherence with, for example, use of written asthma action plans and medical procedures, such as spirometry. 28In the present study, the number of registered spirometry tests was very low.A single measure of lung function may not provide a true estimate of an individual's risk of obstructive disease later. 29However, repeated measurements may reveal a persistent reduction or rapid decline in lung function over time, either of which is more likely to be associated with an increased risk of chronic obstructive pulmonary disease in early adulthood.It is therefore important that adolescents and young adults with asthma have regular follow-up visits, even if their asthma is mild. 8 recent Swedish observational cohort study found that 1 of 5 adult patients with severe asthma had visited specialist care because of asthma during the course of a year. 30Furthermore, they showed that one-third of patients with asthma, irrespective of severity, had visited primary care.The authors discussed that many patients probably did not have regular visits but were managed by prolonged prescription of asthma medication through indirect consultations.Our results showed for all asthma phenotypes low numbers of dispensed asthma medications, that very few had dispensed regular asthma medications, and that more than two-thirds among those with persistent asthma had uncontrolled asthma at the 24-year follow-up.This is unsatisfactory, because the long-term goal of asthma management is to achieve control of symptoms through, for example, treatment.Previous studies have shown that adherence to ICS is poor, leaving patients exposed to the risks of SABA-only treatment. 31,32According to the GINA guidelines' new recommendations, treatment of asthma with SABA alone is no longer suggested for adults and adolescents. 33This supports the importance of consultation and increased understanding of asthma and asthma management. 34n the present study, among those with persistent asthma, high daily doses of ICSs indicated increased odds of having had a consultation after age 18 years.However, almost two-thirds of these young adults had no consultation after age 18 years.Severe asthma also indicated increased odds of having had a consultation, but the adjusted model did not show a significant association.Results from a recent cohort showed that the proportion of children with severe asthma decreased steadily with increasing age, and approximately half of those with severe asthma in childhood resolved during adolescence. 35However, a recent cross-sectional study discussed that severe asthma symptoms, poor lung function, and higher airway hypersensitiveness in childhood are predictors of persistence of childhood asthma to adulthood. 36These groups should therefore, as suggested in guidelines, have more frequent monitoring during and after transition to adult health care.

Strengths and Limitations
Important strengths of the present study include the prospective and population-based design of the BAMSE cohort, and the large and well-characterized study sample.Another strength is the use of solid and unique data through linkage to mandatory Swedish health registries, with high quality and coverage.The national Swedish Prescribed Drug Register provides complete data on the number of individuals exposed to dispensed prescribed medications in the Swedish population. 23The regional VAL database has approximately 85% coverage of all diagnoses in primary care, more than 90% coverage of utilization in specialist care, and more than 99% coverage of hospital care. 37nformation gathered via health care registers may theoretically also have limitations.There is a potential bias in overreporting or underreporting diagnoses and a risk of variability in data quality. 38However, the sensitivity analyses with the wider ICD-10 codes showed comparable results, with a slightly improved mean number of consultations both before and after age 18 years.
The study definition of persistent asthma could be discussed, because symptoms vary over time and in intensity.However, the same results were seen for all asthma phenotypes.
Our results are based on the Swedish population and the generalizability may be questioned because health care systems vary between countries; reimbursement systems, modes of payment, and the uses of specialist services also vary. 21However, the transition from pediatric to adult health care is an international concern, and we believe that our results can be transferred to other countries and populations.

CONCLUSIONS
Almost two-thirds of the young adults with persistent asthma had not had a follow-up visit after age 18 years.For all asthma phenotypes, health care consultations were fewer than recommended in guidelines, and decreased after the transition.The dispensations of asthma medications decreased after the transition, even for the participants with severe asthma.For all asthma phenotypes, almost no one had dispensed regular asthma medications during the 8-year period.This study shows that there is a gap between asthma guidelines and actual management.Increased adherence to current guidelines is required when planning for optimal care of adolescents and young adults, including their transition to adult health care.

FIGURE 2 .
FIGURE 2. Number of consultations 4 years before and after age 18 years, divided by level of care, among young adults with persistent asthma (n ¼ 147).*Consultations in both primary and specialist care were merged into specialist care (n ¼ 4).

FIGURE 3 .
FIGURE 3. The mean number of yearly consultations among young females (n ¼ 79) and males (n ¼ 68) with persistent asthma and the prevalence of asthma during this time period.**The prevalence of current asthma is plotted on the basis of prevalence at the 12-, 16-, and 24-year follow-ups. 15

years before and after 18 years of age
FIGURE 1. Study flowchart of the study population (n ¼ 1808) linked to data sources for asthma-related health care consumption and pharmacological dispensation.BAMSE, Barn/Child, Allergy, Milieu, Stockholm, Epidemiology.

TABLE I .
Number of consultations and level of care, 4 y before and after age 18 y, in relation to asthma at the 16-and 24-y follow-ups Fulfilling at least 2 of the following 3 criteria: symptoms of wheeze and/or breathing difficulties in the last 12 mo, ever doctor's diagnosis of asthma, and/or asthma medicine occasionally or regularly in the last 12 mo.†Fulfilling the definition of current asthma at both the 16-and 24-y follow-ups.zP values obtained using tests of proportions and Wilcoxon signed-rank test indicated differences between 4 y before and after age 18 y for visits and level of care. *

TABLE II .
Number of consultations and level of care, 4 years before and after age 18 y, in relation to asthma phenotypes at the 16-y follow-up

TABLE III .
Asthma phenotypes at the 16-y follow-up in relation to 1 consultation after age 18 y among young adults with persistent asthma (n ¼ 147) *Adjusted for sex and socioeconomic status.
3,16; (2) Taken cortisone tablets dissolved in water for asthma or respiratory symptoms 3 d in a row; (3) Sought acute medical care because of respiratory symptoms; and (4) Impaired lung function, FEV 1 below 80% of predicted.

TABLE IV .
Number of dispensed asthma medicines, 4 y before and after age 18 y, in relation to asthma phenotypes at the 16-y follow-up *A combination of asthma and IgE sensitization to inhalant allergens (cat, dog, horse, and/or house-dust mite, timothy grass, birch, mugwort, and/or mold).

TABLE V .
Number of dispensed asthma medicines, 4 y before and after age 18 y, in relation to asthma at the 16-and 24-y follow-ups