Dry-powder inhaler use in primary school-aged children with asthma: a systematic review Authors: Stephanie L. Kuek, Nicole X. Wong, Stuart Dalziel, Lee Hatter, Louise Fleming, Andrew Bush, Richard Beasley, Shivanthan Shanthikumar Published: 2024 ABSTRACT ObjectiveAsthma is the most common chronic disease among children. Dry-powder inhalers (DPIs) are effective for medication delivery in adults and adolescents, and provide a lower environmental footprint and more portability than a metered dose inhaler (MDI) with a spacer. They require a specific technique, and it is necessary to ascertain whether they can be used in younger age groups. We aimed to assess evidence regarding whether primary school-aged (5–11 years) children can use DPIs with adequate technique during both stable and acute asthma.DesignOvid MEDLINE, Embase and PubMed were searched for studies assessing DPI use among children aged 5–11 years with asthma or wheeze. Results of studies were synthesised by study design and outcome measure.Results38 studies were identified for analysis. These were analysed in three groups: 1) 25 studies assessing children's ability to use DPIs by peak inspiratory flow measurement, 2) 10 studies assessing children's ability to use DPIs by physician assessment, and 3) three studies measuring the efficacy of DPIs compared with the gold standard (MDI with spacer). Five studies included children during acute exacerbations.ConclusionThe majority of primary school-aged children have the ability to use a DPI with adequate training, support and practice. Some younger children may have difficulties, and clinician assessment and ongoing review is crucial in determining which children are likely to benefit from a DPI. Consistent correct use and adherence remain as challenges, but these are also issues with an MDI plus spacer and do not appear to be significantly worse with DPIs. Evidence of the use of DPIs during acute illness is limited; more studies are required in this setting. Introduction Asthma is the most common chronic disease among children and also represents the leading cause of disease burden among primary school-aged children [1–3]. Outcomes remain poor, with childhood asthma being associated with significant morbidity, representing a leading cause of hospital admission [4], quality-of-life impairment [5] and irreversible lung function impairment [6]. Even though most asthma deaths are preventable with appropriate management, children continue to die of asthma around the world [7, 8]. Medication regimens predominantly target bronchodilation and reducing airway inflammation, and the majority of patients can be adequately managed with inhaled medications for both symptom prevention and relief [7]. Inhaled asthma medications can be delivered by a number of devices, including metered dose inhalers (MDIs) with an age-appropriate spacer, and dry-powder inhalers (DPIs). Correct inhaler technique is crucial for effective medication delivery. DPIs require users to inhale “quickly and deeply” to generate appropriate inspiratory flow to activate the device. This varies depending on the internal resistance and other characteristics of the different devices used in clinical practice. While there is good evidence that most adolescents and adults can use DPIs for both reliever and preventer therapy [9], it is less clear whether there is adequate DPI technique to ensure that a DPI is equivalent to an MDI with a spacer in primary school-aged children. When examining national guidelines, some report that a DPI and an MDI with a spacer are equivalent in primary school-aged children [7, 10–12], whereas others report that an MDI with a spacer is preferred in this age group [13]. The main advantages of an MDI with a spacer include ease of use, with only regular tidal breathing required for children administered medication by caregivers. There is also instant feedback from medication administration, with an audible “puff” from propellants, and audible feedback from some spacers with breathing. Component parts can be recycled. Advantages of DPIs include no need for a spacer and no need for hydrofluorocarbon propellants, resulting in a markedly reduced carbon footprint compared with the MDI in current use [14, 15]. Most devices have dose counters; ideally, these should be standard for all inhaler devices. There is a need to establish whether the use of DPIs in primary school-aged children with asthma is appropriate, especially given the push to consider the environmental impact of asthma medications [16]. If DPIs are prescribed to primary school-aged children, and a significant proportion cannot use them appropriately, this could result in additional poor outcomes for many children. In particular, it is important to determine if DPIs can be safely used during acute asthma exacerbations. In this systematic review, we aimed to assess evidence regarding whether primary school-aged (5–11 years) children can use DPIs with adequate technique during both stable and acute asthma. Methods Overview The research question was “can children aged 5–11 years with asthma/wheeze effectively use DPIs with adequate technique?” The study was conducted in accordance with the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) standards [17]. We conducted a literature search through the electronic databases Ovid MEDLINE, Embase and PubMed. Our search included the keywords “respiratory tract diseases” or “bronchial diseases” or “asthma” or “wheeze” combined with “powder” or “DPI” or the names of commonly used DPI device names and the relevant age range (supplementary table S1). This review was not registered. Selection of studies Two authors (SLK and NXW) independently reviewed the titles, abstracts and full texts for inclusion using Covidence. Covidence and EndNote software were used to exclude duplicates. Any disagreements were resolved by third-party adjudication (SS). Reference lists of included papers were screened for any additional inclusions. All authors reviewed the final list for inclusion. Studies were included if they had (1) a population or subgroup of children aged 5–11 years or a paediatric population with a mean age of <10 years and at least five patients within the 5–11 year age range, and (2) a direct or indirect measure of DPI technique. All DPI device types were included. Studies were excluded if their inclusion criteria filtered for children more likely to be able to use DPIs effectively; this included children who had demonstrated adequate DPI technique, previous response to DPI, ability to use a peak inspiratory flow (PIF) measurement device or ability to perform accurate spirometry. Studies assessing medications that are no longer included in the Global Initiative for Asthma (GINA) 2022 recommendations, such as sodium cromoglicate, were also excluded [7]. Studies measuring efficacy were included only if they had a direct comparator of the same medication with the gold standard, an MDI with a spacer; they were excluded if there was another treatment that might mask the effect. Foreign-language studies were excluded if an English translation was not available. Data extraction and quality assessment Data were extracted independently by two reviewers (SLK and NXW): author, title, journal, year of publication, age range, mean age, country of study, study design, measure of efficacy, medication (if relevant), inhaler type (if relevant), total study size (and number in relevant age group), inclusion and exclusion criteria and relevant results. Relevant results varied depending on study type; they comprised PIF, peak expiratory flow (PEF), forced expiratory volume in 1 s (FEV1), symptom scores, physician assessment scores and other. Studies were assessed for risk of bias using the Revised Cochrane Risk-of-Bias tool for Randomized Trials (RoB 2) [18] or the JBI Critical Appraisal checklist for Analytical Cross Sectional Studies [19]. Data synthesis Studies were tabulated and grouped for synthesis by study design and outcome measure. Three groups were identified: (1) studies assessing children's ability to use DPI devices by PIF measurement, (2) studies assessing children's ability to use DPI devices by physician assessment and (3) studies measuring the efficacy of DPIs compared with the gold standard (MDI with spacer). With regard to PIF measurement, there is debate about the minimum PIF required for different DPIs [20, 21]. Therefore, the PIF results were extracted from each study and reported against both the adequate PIF as defined in the study and the generally accepted adequate PIF from the literature, where available. Results A PRISMA diagram summarising the numbers of included studies and reasons for exclusion is shown in figure 1. On initial screening, 1247 studies were identified; 528 of these were selected for full-text review. Papers included for analysis. 38 studies were included for data extraction, with 25 in group 1 (PIF measurement), 10 in group 2 (physician assessment) and three in group 3 (comparison with the gold standard). Key findings are summarised below, with more detailed findings from individual studies in tables 1–4. TABLE 1Studies assessing children's ability to use dry-powder inhalers (DPIs) by measuring peak inspiratory flow (PIF) through the deviceStudyDesignQuality-assessment toolPopulation; n/N (included/total)Age range of included population (mean), yearsDevice(s) usedPIF in L·min−1 (mean±sd unless otherwise stated)#Comments Adachi 2006 [22] Cross-sectional: In-Check™used to assess techniqueJBI Checklist passHealthy children; 28/575–6 (subgroup of 3–6)DiskusNS. 28/28 5–6-year-olds performed PIF >30Mean PIF rate via Diskus versus Turbuhaler for total cohort was 78.0±17.9 L·min−1 versus 64.5±14.3 L·min−1. Weak correlation with heightTurbuhalerNS. 21/28 5–6-year-olds performed PIF >60 Agertoft 1998 [43] Randomised trial: modified pneumatograph used to measure PIF, comparing levels of trainingJBI Checklist passAsthma, stable; 48/724–5 (subgroup of 3–5)Turbuhaler4–5 years: •Basic training only: 33.2•Basic+individual training: 46.4•Individual training+practice: 55Flow improved with additional training and period of practice in the subgroup of 4–5-year-olds. Most 3-year-olds were unable to reach adequate flow, with no difference post trainingNote: the minimal flow required was 30 L·min−1; the ideal was >55 L·min−1 Agertoft 1999 [36] Randomised crossover trial assessing DPI and MDI. PIF measured by Vitalograph, and drug deposition on inhaler assessedJBI Checklist passAsthma, stable; 1983–15 (9)Turbuhaler5 years: 89% >50Higher in all older age groups (graph; numbers not specified)PIF was higher among children who had used Turbuhaler previously than it was among healthy children. Fewer children aged <5 years were able to generate sufficient flow Amirav 2005 [23] Cross-sectional: In-Check™ used to mimic internal resistance of device to measure PIFJBI Checklist passAsthma, stable; 100/2233–15 (9.1)Turbohaler64±18Turbohaler “high resistance” and Diskus “low resistance”. PIF specified for “experienced” groupRemainder were “inexperienced” group with PIF not specified: 123 patients, mean age 5.5±1.9 years. For the inexperienced group, minimal PIF >30 L·min−1 for Diskus and 60 L·min−1 for Turbuhaler was achieved in 4- and 9-year-olds, respectively. Positive relationship between age and PIFDiskus77±14 Azouz 2015 [44] Randomised crossover study; training via In-CheckJBI Checklist passAsthma, stable; 23/1006–11 (8.57) (paediatric subgroup)Spiromax77.99±17.64Improved with training by 14.18 and 20.03 L·min−1Turbuhaler69.46±16.18 Azouz 2015 [35] Cross-sectional: PIF measured with MicroLoop spirometerJBI Checklist passAsthma, stable; 16/985–17 (8.8) (children with asthma subgroup)Aerolizer71.4±21.5Children had lower PIF than adults with asthmaDiskus53.3±24.2Turbuhaler44.8±16.0Easyhaler45.5±13.2 Bisgaard 1998 [24] Cross-sectional: PIF and particle sizeJBI Checklist passAsthma, stable; 18/368 (8.4)Also 4-year-old subgroupDiskus (fluticasone 250 μg)Home values: •4 years: 57±19 to 80±22•8 years: 97±29 to 121±30Lab values: •4 years: 70±23, 8 years: 105±14Diskus: higher proportion of coarse particles/improved dose consistency across age groups (according to proportion of dose emitted at each particle size)Two 4-year-olds were excluded as unable to perform PIF manoeuvre sufficiently to measureTurbuhaler (budesonide 200 μg)Lab values: •4 years: 53±23•8 years: 76±10 Bronsky 2004 [25] Cross-sectional: PIF measured via spirometry+adaptorJBI Checklist passAsthma, stable; 32/657–11 (9.6) (paediatric subgroup)Aerolizer (formoterol)100±5 (91% >60)Loading Aerolizer device reduced mean PIF rate from 157±9 L·min−1. No correlation between asthma severity and PIF rate Cooper 2020 [37] Crossover studyJBI Checklist passAsthma, stable; 20/68 20/68 (paediatric subgroups) 4–7 (5.6)50.6±11.13Inhub had slightly higher resistanceInhub Diskus 66.8±16.328–11 (9.7)Inhub75.4±9.53Diskus100.6±10.07 Desager 2006 [38] Randomised trial: In-Check™ used to assess techniqueJBI Checklist passAsthma, stable; 306–16 (9.5)AerolizerMedian 120 (range 70– >120)Inspiratory flow exceeded 60 L·min−1 in all but one child (who was using the Turbuhaler)DiskusMedian 110 (range 70– >120)TurbuhalerMedian 85 (range 50– >120) Iqbal 2003 [26] Cross-sectionalJBI Checklist passAsthma, stable and healthy children; 17/615 (subgroup of 2–5)ClickhalerAsthmatic: 60.5Non-asthmatic: 54.8Reports >15 L·min−1 as adequate. 62.5% of 3-year-olds and 100% of 4-year-olds achieved a PIF of >15 L·min−1 Jogi 2019 [39] Randomised trial: Easyhaler versus DiskusJBI Checklist passAsthma, stable; 60/2274–11 (7.5) (paediatric subgroup)Easyhaler (salmeterol/fluticasone)54.5±16.6Diskus (salmeterol/fluticasone)67.9±22.6 Kamps 2004 [46] Cross-sectional: In-Check™ used to assess technique (home environment)JBI Checklist passAsthma, stable or acute; 29/58 29/58 4–15 (8.9) Accuhaler Initial: 110.7±28.9At home: 109.9±21.8After 4 weeks: 118.3±23.0 Both cohorts showed an increase in PIF at the second visit after a 4-week trial at home. Mean maximal decrease in PIF of 11 L·min−1 with increase in asthma symptoms. Children ≤7 years had greater variability of PIF. PIF correlated with age and height. Asthma symptoms associated with decrease in PEF and PIF4–15 (8.0)TurbohalerInitial: 82.0±22.6At home 82.7±27.3After 4 weeks: 86.4±21.0 Malmstrom 1999 [27] Cross-sectional: PIF measured with pneumotachographJBI Checklist passAsthma, stable; 1204–16 (9.1)Easyhaler (salbutamol 200 μg)56±15PIF correlated with age, height and PEF. Additional bronchodilation assessment: similar bronchodilatory effects of salbutamol via non-primed spacer versus Easyhaler Manuyakorn 2010 [28] Cross-sectional: In-Check™ used to assess techniqueJBI Checklist passAsthma, stable; 194/2595–12 (NS) (subgroup of 5–18)Turbuhaler90% of 5–6-year-olds achieved ≥6096.6% of 7–12-year-olds achieved ≥60100% of 5–12-year-olds achieved >30PIF correlated with age for both devices. 30 L·min−1 was considered minimum adequate PIF and 60 L·min−1 optimal for both devicesAccuhaler100% of 5–12-year-olds achieved >60 Negro 2001 [29] Cross-sectional: PIF measured with pneumotachographJBI Checklist passAsthma, stable; 16/89 (paediatric subgroup)7–13 (9.2)Pulvinal39.0±10.0All patients achieved PIF >20 L·min−1, considered sufficient. No correlation between PIF rate and FEV1 parameters Nielsen 1997 [30] Cross-sectional: PIF measured via pneumotachographJBI Checklist passAsthma, stable; 1263–10 (NS)AerolizerWhole cohort: 104>8 years: 120All children (except one) aged ≥5 years performed PIF >60 L·min−1¶ Nikander 2011 [45] Cohort: PIF measured via pneumotachograph, assessed adherence over timeJBI Checklist passAsthma, stable; 1155–10Turbuhaler (budesonide)5–6 years: start 53±7, end 63±127–8 years: start 64±17, end 73±319–10 years: start 63±7, end 69±18All individual PIF values except two aged 6 and 8 years exceeding competence level, defined as 40 L·min−1. Adherence in last 45 days of trial was 57.4% Parry-Billings 2003 [31] Cross-sectional: inspiratory flow measured by calibrated flow rate monitorJBI Checklist passAsthma, stable; 20/57 21/57 6–8 Clickhaler 40.5±16.0, 50% achieved ≥30 L·min−1 on both of two attempts 2.3% failed to achieve a PIF of >30 L·min−19–11 (subgroups of 6–17)62.5±12.1, 86% ≥30 L·min−1 on both of two attempts Pedersen 1990 [32] Cross-sectionalJBI Checklist passWheeze, acute; 344–13 (7.4)Turbuhaler4–5 years: 24 (range 14–36)6–13 years: 39 (range 30–54)9/15 <6 years inspiratory flow <28 L·min−12/19 ≥6 years inspiratory flow <28 L·min−1One third of three studies reported; study 2 found a correlation with PIF and age in healthy children Raissy 2006 [40] Randomised crossover study: PIF measured via SpiroStar spirometerJBI Checklist passAsthma, stable; 31/605–12 (NS) (subgroup of 2–12)AerolizerNS; 30/30 5–12-year-olds had PIF >30NS; 16/30 5–12-year-olds had PIF >60Mean for cohort of 2–12-year-olds: •Aerolizer: 55.7±23.7 L·min−1•Diskus: 48.4±18.9 L·min−1•Turbuhaler 37.5±14.8 L·min−1Mean PIF of Aerolizer>Diskus>TurbuhalerDiskusNS; 23/27 5–12-year-olds had PIF >30NS; 13/28 5–12-year-olds had PIF >60TurbuhalerNS; 25/30 5–12-year-olds PIF >30NS; 4/30 5–12-year-olds had PIF >60 Ruggins 1993 [41] Randomised crossover study: PIF measured using mercury pneumotachographJBI Checklist passAsthma, acute; 514–13 (9)Rotahaler (salbutamol 100 μg)109.5±45 first study period; 113.5±44 second study periodAssessed both PIF and device “failure” as measured by inability to fire Autohaler or incomplete emptying of Rotacap. Autohaler 1/100 failure, Rotahaler 26/100 failureAll children <6 years unable to activate Rotahaler on at least one of two attempts Tantibhaedhyangkul 2003 [33] Cross-sectional: In-Check™+Turbutester used to assess techniqueJBI Checklist passWell children; 272/4685–7 (subgroup of 3–7)In-Check5 years: 1166 years: 1237 years: 127Performance improved for children >5 years, correlating with weight of >20 kg and height >113 cm. PIF via In-Check was higher than with Turbutester. 90% of children aged >5 years were assessed as having the ability to use the TurbuhalerTurbuhaler5 years: 726 years: 767 years: 77 Vogelberg 2004 [34] Cross-sectional: PIF measured using pneumotachographJBI Checklist passAsthma, stable; 1364–11 (7)NovolizerMean of three attempts: 69±18PIF increased with age until 8 years of age (no increases seen above this limit). Five out of 136 were unable to overcome the trigger threshold, all aged 4–5 years von Berg 2007 [42] Randomised crossover, PIF measured using pneumotachographJBI Checklist passAsthma, stable; 466–11 (8.5)Novolizer88.3±18.8Three attempts – first attempt in each series 3–5% lower than the mean. One 6-year-old girl unable to activate Novolizer on three attemptsTurbuhaler65.4±12.8#: PIF for effective use is 30 L·min−1 for Diskus, 30–60 L·min−1 for Turbuhaler, 30 L·min−1 for Accuhaler and 60 L·min−1 for Aerolizer. Autohaler activated by minimum of 30 L·min−1. ¶: Additional cohort had PIF assessed using Aerolizer with formeterol versus placebo following exercise challenge (mean of group >10 years, not included). NS: not specified; MDI: metered dose inhaler; PEF: peak expiratory flow; FEV1: forced expiratory volume in 1 s. Studies assessing children's ability to use dry-powder inhalers (DPIs) by measuring peak inspiratory flow (PIF) through the device •Basic training only: 33.2 Basic training only: 33.2 •Basic+individual training: 46.4 Basic+individual training: 46.4 •Individual training+practice: 55 Individual training+practice: 55 •4 years: 57±19 to 80±22 4 years: 57±19 to 80±22 •8 years: 97±29 to 121±30 8 years: 97±29 to 121±30 •4 years: 70±23, 8 years: 105±14 4 years: 70±23, 8 years: 105±14 •4 years: 53±23 4 years: 53±23 •8 years: 76±10 8 years: 76±10 •Aerolizer: 55.7±23.7 L·min−1 Aerolizer: 55.7±23.7 L·min−1 •Diskus: 48.4±18.9 L·min−1 Diskus: 48.4±18.9 L·min−1 •Turbuhaler 37.5±14.8 L·min−1 Turbuhaler 37.5±14.8 L·min−1 #: PIF for effective use is 30 L·min−1 for Diskus, 30–60 L·min−1 for Turbuhaler, 30 L·min−1 for Accuhaler and 60 L·min−1 for Aerolizer. Autohaler activated by minimum of 30 L·min−1. ¶: Additional cohort had PIF assessed using Aerolizer with formeterol versus placebo following exercise challenge (mean of group >10 years, not included). NS: not specified; MDI: metered dose inhaler; PEF: peak expiratory flow; FEV1: forced expiratory volume in 1 s. Group 1: studies assessing children's ability to use DPI devices by PIF measurement 25 studies assessed children's ability to generate sufficient PIF to use DPIs effectively (table 1). 22 studies involved children with asthma, two involved healthy children and one involved both groups. These were published between 1990 and 2020, with a variety of study designs. 14 cross-sectional studies assessed PIF at a point in time, predominantly in real-world scenarios [22–35]. Seven were randomised trials comparing more than one device, six of which had a crossover period [36–42]. Four studies assessed the impact of training or time on children's PIF [43–46]. Most studies used pneumotachograph devices to measure PIF. Some had varying internal resistance settings and some were devised specifically to mimic the internal resistance of a specific device [28, 33]. Multiple devices were evaluated (supplementary table S2). The definition of adequate PIF varied between devices assessed, with most studies using a minimum sufficient PIF of between 30 L·min−1 and 60 L·min−1. There was even variation between the definition of adequate PIF in studies evaluating the same device. For example, the optimal PIF for the Turbuhaler (a DPI with medium to medium-high resistance) is generally considered to be 55–60 L·min−1 [22, 23, 47, 48]; however, some studies reported effective drug delivery at lower flows of 28–30 L·min−1 [28, 32, 43]. We did not identify any studies assessing PIF for some DPIs, including the Breo Ellipta, Breezhaler, Flexhaler and HandiHaler. The most commonly assessed device was the Turbuhaler (or Turbohaler), which was the focus of 15 studies involving 1181 participants. The majority of studies (11/15) found that the mean PIF among participants was >60 L·min−1, although some studies reported lower acceptable thresholds of >28 or >30 L·min−1. The proportion of participants who demonstrated sufficient PIF (post training and based on the minimum threshold defined in each study) ranged between 54.2% and 98.3% (table 2). Fewer children were able to generate a PIF of >60 L·min−1 than could generate a PIF of 30 L·min−1, such as in Raissy et al. [40], in which 13.3% of 5–12-year-olds were able to generate the “optimal” PIF of >60 L·min−1, compared with 83.3% able to generate the “acceptable” PIF of >30 L·min−1. 11 studies reported the sample size and proportion of participants with adequate PIF, and 792 of the 871 participants (90.9%) in these studies generated adequate PIF. When limited to only studies reporting the proportion of participants with a PIF >55–60 L·min−1 (the most commonly used threshold for optimal Turbuhaler use), 321 of the 398 participants (80.7%) generated sufficient PIF. One study examined patients during a 28-day period and found that 28 out of 29 participants maintained a PIF of >30 L·min−1 during the entire study period [46]. They also reported that the mean reduction of PIF during periods of severe asthma symptoms, when compared with baseline values, was 11.0 L·min−1. A further study monitored PIF over the course of 18 months and found that 113 out of 115 participants maintained a PIF of >40 L·min−1 for the entire study period [45]. TABLE 2The proportion of children who could generate adequate peak inspiratory flow (PIF)StudyReported adequate PIF threshold (L·min−1)Adequate PIF (n)Total participants (n)Proportion with adequate PIF (%)Comments Turbuhaler Adachi 2006 [22]60212875 Agertoft 1998 [43]5511Before training5554After in-clinic training5584After home training Agertoft 1999 [36]5089 Amirav 2005 [23]606810068 Azouz 2015 [44]302323100After training Azouz 2015 [35]3014168860131681 Desager 2006 [38]60293097 Kamps 2004 [46]30282997Adequate technique over 28 days of twice-daily assessment Manuyakorn 2010 [28]301931949930 L·min−1=minimum PIF601861949660 L·min−1=optimal PIF Nikander 2011 [45]4011311598Tested over 18-month trial Pedersen 1990 [32]28233468During acute wheeze Raissy 2006 [40]3025308330 L·min−1=acceptable PIF604301360 L·min−1=optimal PIF Tantibhaedhyangkul 2003 [33]3025527294 Diskus Adachi 2006 [22]302828100 Amirav 2005 [23]30100100100 Azouz 2015 [35]30131681 Desager 2006 [38]603030100 Raissy 2006 [40]3023278560132846 Other Azouz 2015 [35]301616100Aerolizer30151694Easyhaler Bronsky 2004 [25]60293291Aerolizer Desager 2006 [38]603030100Aerolizer Iqbal 2003 [26]15161794Clickhaler Kamps 2004 [46]302929100Accuhaler; adequate technique over 28 days of twice-daily assessment Malmstrom 1999 [27]2811612097Easyhaler Manuyakorn 2010 [28]30194194100Accuhaler; 30 L·min−1=minimum PIF60194194100Accuhaler; 60 L·min−1=optimal PIF Negro 2001 [29]201616100Pulvinal60868799Aerolizer Parry-Billings 2003 [31]154141100Clickhaler Raissy 2006 [40]303030100Aerolizer60163053Aerolizer Ruggins 1993 [41]Device emptying4040100Autohaler; during acute wheezeDevice emptying314078Rotahaler; during acute wheeze Vogelberg 2004 [34]35104104100Novolizer von Berg 2007 [42]354646100Novolizer The proportion of children who could generate adequate peak inspiratory flow (PIF) The Diskus (low-resistance DPI) was the second most commonly assessed device, being evaluated in eight studies involving 328 participants. All eight studies reported a mean PIF of >30 L·min−1. Five studies reported the proportion of participants with adequate PIF (predominantly defined at >30 L·min−1), and 194 out of 201 participants (96.5%) were able to generate adequate PIF (table 2). Studies evaluating other devices (listed in supplementary table S2) demonstrated that the majority of children could generate adequate PIF [25–29, 31, 34, 35, 37–42, 44, 46]. The range of participants who generated an adequate PIF across 13 studies was 77.5–100% (table 2). 11 studies involved simultaneous assessment of two or more devices. Younger children were more likely to generate an adequate PIF when using lower-resistance devices (e.g. Diskus) than when using higher-resistance devices (e.g. Turbuhaler) [22–24, 35, 38, 40]. In five studies that compared the Turbuhaler with the Diskus, two showed that a similar proportion of children could generate an adequate PIF with either device [35, 38], while three showed that a higher proportion of children could generate a sufficient PIF with the Diskus [22, 23, 40]. Similarly, in five studies that compared inhalers other than the Diskus with the Turbuhaler, three showed that a similar proportion of children could generate a PIF adequate for the tested devices [28, 38, 46], whereas two showed that higher proportions of children could use the Aerolizer [35, 40] and one showed that a higher proportion could use the Easyhaler (medium to medium-high resistance) [35]. Studies that compared PIF between younger primary school-aged children and older children tended to show lower values in the younger children [24, 27–29, 31, 36, 45, 46]. The definition of “younger” and “older” varied between studies; for example, Bisgaard et al. [24] found a higher PIF in 8-year-olds than in 4-year-olds using Turbuhaler and Diskus, and Cooper et al. [37] found a higher PIF in 8–11-year-olds than in 4–7-year-olds using Inhub and Diskus. Manuyakorn et al. [28] found that 90% of 5–6-year-olds and 96.6% of 7–12-year-olds were able to generate a PIF of >60 L·min−1, and 100% of 5–12-year-olds could generate a PIF of >30 L·min−1. Two studies included measurements pre- and post-device Turbuhaler training, and both demonstrated improvements in PIF with training [43, 44]. Acute asthma Only three studies [32, 41, 46] assessed children during acute wheeze. Pedersen et al. [32] assessed 34 asthmatic children using the Turbuhaler during acute wheeze on a visit to the outpatient department, reporting a minimum sufficient PIF of 28 L·min−1. They found that six out of 15 children aged 4–5 years and 17 out of 19 children aged 6–13 years generated a PIF of >28 L·min−1 (mean PIF 24 (range 14–36) L·min−1 among 4–5-year-olds and 39 (range 30–54) L·min−1 among 6–13-year-olds reaching target PIF). Ruggins et al. [41] examined the Rotahaler and breath-actuated MDI Autohaler in children at least 2 h post salbutamol nebulisation during hospitalisation for acute asthma. They reported a mean±sd PIF of 109.5±45 L·min−1 on initial assessment and 113.5±44 L·min−1 on second assessment; the lowest PIF recorded was 30 L·min−1. Despite this, 20 patients, including all 11 children younger than 6 years, failed to empty the Rotacap on at least one occasion. Kamps et al. [46] assessed children across 4 weeks at home using Turbohaler and Accuhaler. 30 L·min−1 was defined as adequate PIF for Turbohaler in this study. They found that both PIF and PEF reduced as symptoms of asthma increased, reporting morning PIF reductions of 3.1, 8.3 and 11.0 L·min−1 for parental report of mild, moderate and severe asthma symptoms, respectively (four-point Likert scale). There was a greater variability in PIF among children aged <7 years and with acute symptoms, but the majority of children still generated a PIF of >30 L·min−1. Group 2: studies assessing children's ability to use DPI devices by physician assessment 10 studies, published between 1994 and 2021, used physician assessment to assess children's ability to use DPIs, and involved 1848 participants (table 3). Nine studies used a structured questionnaire for physician assessment [49–57], and one captured drug delivery amount on a filter [58]. All studies included children with stable asthma or stable recurrent wheeze. Multiple devices were used: Turbuhaler, Diskus, Rotahaler, Easyhaler, Ellipta and Diskhaler. Six studies assessed more than one device, including four studies that also assessed an MDI with a spacer. TABLE 3Studies assessing children's ability to use dry-powder inhalers (DPIs) by physician assessmentStudyDesignQuality-assessment toolPopulation; n/N (included/total)Age range of included population (mean), yearsDevice(s) usedAssessmentFindings/comments Alshehri 2003 [50] Cross-sectional survey to assess inhaler techniqueJBI Checklist passAsthma, stable; 2715–12 (NS)Turbuhaler, Diskus, MDIClinician assessment of technique with structured questionnaireAll steps carried out correctly: one out of four patientsOver half of the steps performed correctly: two out of three DPI patients and one out of three MDI patientsMDI direct to mouth, no spacer Arumugom 2017 [51] Pilot open-label randomised controlled trial to compare verbal with video demonstration to improve inhaler techniqueJBI Checklist passAsthma, stable; 286–12 (8.2)RotahalerClinician assessment of technique with structured questionnaireAll steps carried out correctly: •Verbal instruction group: 2/14 (14.28%) at baseline, 5/14 (35.71%) after 1 month of follow-up•Video demonstration group: 8/14 (57.14%) at baseline, 13/14 (92.86%) at 1 month of follow-upVideo demonstration and re-education to correct errors was effective to improve Rotahaler technique Bisgaard 1994 [58] Cross-sectional assessment of inhaler techniqueJBI Checklist passRecurrent wheeze, stable; NS/1656–8 years (subgroup of group aged 8 months–8 years)TurbuhalerDrug delivery on filterChildren >6 years obtained median 55% nominal dose. Cannot rely on predictable dose delivery in young childrenPoint in time only, cannot comment on practice, and concern when child upset/unwell Gálffy 2013 [52] Open, uncontrolled, non-randomised study to evaluate efficacy, safety and patient satisfactionJBI Checklist passRecurrent wheeze, stable; 139/2193–11 (subgroup of 3–17)EasyhalerClinician assessment of technique with structured questionnaireAll steps carried out correctly: •One demonstration: 64% of children (76% of adolescents, 78% of adults and 70% of elderly people)•First visit: 61.9% of children (65.0% of adolescents)•Second visit: 83.5% of children (88.8% of adolescents) Goren 1994 [53] Open, non-controlled study to determine at what age young asthmatic children could use device after a single instructionJBI Checklist passAsthma, stable; 30/595–6 (subgroup of 3–6)Turbuhaler and DiskusClinician assessment of technique with structured questionnaireAll steps carried out correctly: 10/15 (67%) 5-year-olds, 12/15 (80%) 6-year-olds (0% of 3-year-olds, 43% of 4-year-olds)12/13 5-year-olds and 15/15 6-year-olds had clinical improvement Halverson 2021 [54] Multicentre single-arm stratified open-label 28-day placebo studyJBI Checklist passAsthma, stable; 2225–7, 8–11ElliptaClinician assessment of technique with structured questionnaireCorrect use on first attempt: •Visit 1: 21/88 (24%) 5–7-year-olds and 67/134 (50%) 8–11-year-olds•Visit 2: 81/88 (92%) 5–7-year-olds, 124/133 (95%) 8–11-year-olds5/222 (2%) required a maximum of five attempts to demonstrate correct use; no child demonstrated a critical error on the fifth attempt Hantulik 2015 [55] Open, multicentre non-interventional observational study to validate a questionnaire and examine usage and usabilityJBI Checklist passAsthma, stable; 164/5423–12 (9.6; paediatric subgroup)Easyhaler or others (Diskhaler, Turbuhaler or MDI)Clinician assessment of ease of teaching with structured questionnaire, patient/carer questionnaireVery or fairly easy to teach for 88% with Easyhaler and 71.6% with another deviceNo significant difference between Easyhaler and other devicesSponsored by pharmaceutical company; non-blinded Kamps 2000 [49] Open study, physician assessment of children using their inhalers before and after inhalation instructionsJBI Checklist passAsthma, stable; 46/9517 newly referred29 known asthmatic6–14 (9)5–10 (7)Diskhaler or Turbuhaler, MDIClinician assessment of technique with structured questionnaireAll steps carried out correctly: •New to clinic with no re-instruction: 58% of all, 29% of DPI subgroup•Control group/standardised instruction: 27/29 (93%) of all, 100% of DPI subgroup•23/29 (79%) of pharmacy-instructed, 12/31 (39%) of GP-instructed (pooled, difference independent of inhaler) Sleath 2012 [56] Observational study of provider–patient communication about asthma medication devices during primary care clinic visitsJBI Checklist passAsthma, stable; 71 Diskus; 29 Turbuhaler8–12 (subgroup of 8–16)Diskus, Turbuhaler, MDIClinician assessment of technique with structured questionnaireAll steps carried out correctly: •Turbuhaler: 6/29 (20.7%) 8–12-year-olds (2/9 (22.2%) 13–16-year-olds)•Diskus 17/71 (23.9%) 8–12-year-olds (11/33 (33.3%) 13–16-year-olds)•MDI with spacer: 29/174 (16.7%) 8–12-year-olds (16/76 (21.1%) 13–16-year-olds)Improved technique with improved education – multiple parameters Vichyanond 1994 [57] Open, non-comparative multicentre trial assessing handling technique, change after combined written and verbal instructionJBI Checklist passAsthma, stable; 835–14 (8.7)Turbuhaler and DiskusClinician assessment of technique with structured questionnaireAll steps carried out correctly: •24% after initial written instruction (noted some could not read)•73% after written+verbal instruction•99% after 4 weeks’ active treatmentImproved clinical scores and PEF after 4 weeks. Good device acceptanceNS: not specified; MDI: metered dose inhaler; PEF: peak expiratory flow. Studies assessing children's ability to use dry-powder inhalers (DPIs) by physician assessment •Verbal instruction group: 2/14 (14.28%) at baseline, 5/14 (35.71%) after 1 month of follow-up Verbal instruction group: 2/14 (14.28%) at baseline, 5/14 (35.71%) after 1 month of follow-up •Video demonstration group: 8/14 (57.14%) at baseline, 13/14 (92.86%) at 1 month of follow-up Video demonstration group: 8/14 (57.14%) at baseline, 13/14 (92.86%) at 1 month of follow-up •One demonstration: 64% of children (76% of adolescents, 78% of adults and 70% of elderly people) One demonstration: 64% of children (76% of adolescents, 78% of adults and 70% of elderly people) •First visit: 61.9% of children (65.0% of adolescents) First visit: 61.9% of children (65.0% of adolescents) •Second visit: 83.5% of children (88.8% of adolescents) Second visit: 83.5% of children (88.8% of adolescents) •Visit 1: 21/88 (24%) 5–7-year-olds and 67/134 (50%) 8–11-year-olds Visit 1: 21/88 (24%) 5–7-year-olds and 67/134 (50%) 8–11-year-olds •Visit 2: 81/88 (92%) 5–7-year-olds, 124/133 (95%) 8–11-year-olds Visit 2: 81/88 (92%) 5–7-year-olds, 124/133 (95%) 8–11-year-olds •New to clinic with no re-instruction: 58% of all, 29% of DPI subgroup New to clinic with no re-instruction: 58% of all, 29% of DPI subgroup •Control group/standardised instruction: 27/29 (93%) of all, 100% of DPI subgroup Control group/standardised instruction: 27/29 (93%) of all, 100% of DPI subgroup •23/29 (79%) of pharmacy-instructed, 12/31 (39%) of GP-instructed (pooled, difference independent of inhaler) 23/29 (79%) of pharmacy-instructed, 12/31 (39%) of GP-instructed (pooled, difference independent of inhaler) •Turbuhaler: 6/29 (20.7%) 8–12-year-olds (2/9 (22.2%) 13–16-year-olds) Turbuhaler: 6/29 (20.7%) 8–12-year-olds (2/9 (22.2%) 13–16-year-olds) •Diskus 17/71 (23.9%) 8–12-year-olds (11/33 (33.3%) 13–16-year-olds) Diskus 17/71 (23.9%) 8–12-year-olds (11/33 (33.3%) 13–16-year-olds) •MDI with spacer: 29/174 (16.7%) 8–12-year-olds (16/76 (21.1%) 13–16-year-olds) MDI with spacer: 29/174 (16.7%) 8–12-year-olds (16/76 (21.1%) 13–16-year-olds) •24% after initial written instruction (noted some could not read) 24% after initial written instruction (noted some could not read) •73% after written+verbal instruction 73% after written+verbal instruction •99% after 4 weeks’ active treatment 99% after 4 weeks’ active treatment NS: not specified; MDI: metered dose inhaler; PEF: peak expiratory flow. Overall, the proportion of children completing all steps correctly on initial assessment (without additional training) was low, with the majority of studies finding that <30% of children were able to complete all steps correctly [49–51, 56, 57]. Two studies found that a higher proportion of children could use a DPI on first attempt, with Gálffy et al. [52] reporting that 61.9% of 139 children aged 3–11 years could use the Easyhaler correctly, and Goren et al. [53] reporting that 73.3% of 59 children aged 3–6 years could use the Turbuhaler or Diskus. In the four studies that involved an MDI as a comparator, all reported that an equal or greater proportion of children could successfully complete all steps using a DPI, compared with children using an MDI, with adequate instruction [49, 50, 55, 56]. For example, Sleath et al. [56] found that, among 8–12-year-olds, 20.7% of Turbuhaler users, 23.9% of Diskus users and 16.7% of MDI with spacer users carried out all steps correctly on initial assessment. These numbers remained low for teenagers; among those aged 13–16 years, 22.2%, 33.3% and 21.1% of Turbuhaler, Diskus and MDI-with-spacer users, respectively, carried out all steps correctly [56]. Four studies assessed how training affects ability to use devices correctly [49, 51, 56, 57]. All showed increased correct use with higher-quality training. Correct use also improved when carer assistance and prompting was allowed, and with repeated instruction and practice over time. One study assessed particle deposition on a filter as a surrogate for drug delivery; Bisgaard et al. [58] found that the subgroup of 6–8-year-olds obtained median 55% nominal dose, concluding that predictable dose delivery is unreliable among young children. Group 3: studies assessing efficacy of DPIs by comparison with the gold standard Three randomised studies compared DPI with gold-standard delivery devices; they were published between 1991 and 2008 and involved 106 participants (table 4). Two studies assessed children presenting with acute wheeze or asthma [59, 60] and one assessed children with stable asthma [61]. Two studies compared DPIs with MDIs with spacers, and one study compared DPIs with MDIs with spacers and nebulised salbutamol. There were no significant differences in clinical parameters in any group when comparing DPIs with the gold standard. TABLE 4Studies assessing efficacy of dry-powder inhalers (DPIs) by comparison with the gold standardStudyDesignQuality-assessment toolPopulation; nAge range of included population (mean), yearsDevice usedComparatorMedicationFindings/comments Rufin 1991 [59] Open, randomised, parallel-group studyRoB 2 passAcute wheeze; 224–14 (9)TurbohalerMDI with spacerBudesonideLetter to the editor. No significant difference between treatments when measured sRaw, FEV1 and PEF rate at 15 and 30 min Singh 2001 [61] Open randomised crossover trialRoB 2 passAsthma, stable; 306.1±2.7RotahalerMDI with spacerBeclomethasone, salbutamolNo statistically significant difference between MDI and DPI at most time points when assessing PEF and symptom score Vangveeravong 2008 [60] Randomised controlled trialRoB 2 passAcute asthma (mild to moderate); 545–18 (8.4)EasyhalerNebuliser and MDI with spacerSalbutamolImprovement in all three groups; no statistically significant difference between groups when assessing clinical response by vital signs and physician-assessed asthma scoreRoB 2: Revised Cochrane Risk-of-Bias tool for Randomized Trials; MDI: metered dose inhaler; sRaw: specific airway resistance; FEV1: forced expiratory volume in 1 s; PEF: peak expiratory flow. Studies assessing efficacy of dry-powder inhalers (DPIs) by comparison with the gold standard RoB 2: Revised Cochrane Risk-of-Bias tool for Randomized Trials; MDI: metered dose inhaler; sRaw: specific airway resistance; FEV1: forced expiratory volume in 1 s; PEF: peak expiratory flow. Rufin et al. [59] reported no significant difference in FEV1, PEF or specific airway resistance when comparing treatment with budesonide administered by Turbohaler with an MDI with spacer in 22 children with acute wheeze. Singh et al. [61] reported an open randomised crossover trial comparing Rotahaler with MDI with spacer in 30 children with stable asthma, again finding no difference in PEF and symptom score at most time points. Vangveeravong et al. [60] reported a randomised controlled trial comparing salbutamol delivery between DPI (Easyhaler), MDI with spacer and nebuliser in 54 children with mild to moderate acute asthma. All groups showed clinical improvement post treatment, and there was no statistically significant difference between groups when assessing clinical response by vital signs and physician-assessed asthma score. Discussion Correct and consistent inhaler use is crucial in asthma management [7]. An MDI with a spacer provides excellent medication delivery, but propellants are harmful to the environment and spacers are bulky and not used consistently. DPIs are widely used among adolescents and adults, including during acute exacerbations, and may be an alternative to MDIs with spacers among younger school-age children [9, 62]. Furthermore, there has been a shift towards prescribing anti-inflammatory reliever regimens, including inhaled corticosteroid/formoterol maintenance and reliever therapy in adolescents and adults with asthma, with the evidence to support this approach generated using DPI devices [63–66]. The results of this systematic review show that the majority of primary school-aged children with stable asthma are capable of using a DPI with adequate technique. Overall, the majority of school-age children can generate sufficient PIF to effectively use a DPI. There is a positive correlation between age and PIF in primary school-aged children, with older primary school-aged children demonstrating higher PIF. While some 4–6 year-olds can produce adequate PIF, particular care should be taken in this age group. Some studies have also found DPIs to be equivalent to MDIs in 2–5-year-olds [67]; however, definitions of adequate PIF and results vary, and as such there is insufficient evidence to recommend DPI use among preschool-aged children. Lower-resistance devices may be more suitable for younger children or those struggling to achieve the higher PIF required for higher-resistance devices such as the Turbuhaler. Adequate PIF alone is not sufficient to use a DPI effectively, and additional factors must be considered [21]. The negative pressure generated by a patient's inspiratory effort is key in effective DPI use, and can be affected by a variety of factors including lung volume, mouth seal and muscle strength [20]. Device factors such as varying resistance play a role, with higher-resistance devices, such as the Turbuhaler, generating more turbulent flow and lower PIF for the same pressure change. Most DPIs require a minimum PIF of 30 L·min−1 to be triggered [21], but this varies between devices. While there are generally agreed PIF values at which DPIs function optimally, there is some disagreement as to a set threshold below which the device cannot be used [20]. As such, there was a wide variety in what was reported as “adequate” PIF in the studies reviewed. For example, the clinically effective range for the Turbuhaler is commonly reported as 30–90 L·min−1 [21]. Pedersen et al. [32] reported reduced effect once PIF reached <28 L·min−1, although with small numbers. Other studies describe >55 or >60 L·min−1 [22, 23, 30, 47, 48] as ideal. This variation in the definition of adequate PIF is a significant limitation for interpreting the current available evidence with regard to the research question. Without an agreed adequate PIF, it is difficult to determine the proportion of primary school children who can use a DPI. Alternatively, rather than focusing on PIF as the primary outcome measure, future studies could look at more clinically relevant outcomes such as symptoms, physician assessment, spirometry and, in children with acute wheeze, vital signs such as respiratory rate and oxygen saturation. Physical ability to inhale effectively through a DPI differs from adherence and capacity to do this consistently. In real-world observational studies, technique was found to be generally poor [50, 53, 54, 58]. This issue is not unique to children using DPIs; children using MDIs, as well as adults and adolescents, all had low rates of completing all steps correctly on clinician assessment [49, 52, 56]. Furthermore, there is likely to be a significant difference between what children can be persuaded to do in highly controlled clinical trial settings and what they do in real life, which is likely to be further compounded by what they can do when well and during an acute asthma attack. Adherence monitors are useful in assessing children's adherence to treatment, and more recently mobile direct observation of therapy has been trialled [68]. While challenging from a resource perspective, this may be useful in some cases. Repeated education and assessment of inhaler technique is crucial, regardless of what type of device is chosen. Though most school-aged children are physically capable of generating sufficient PIF and using a DPI when assessed with stable asthma, there remains a minority who cannot. With ongoing significant morbidity and mortality from asthma, it is crucial to identify these children for whom DPIs are not appropriate. Some children are not able to use DPIs even with stable asthma, and this is a major concern regarding what happens during an acute attack. The consequences are potentially catastrophic if a child is given an inhaler that they cannot use, thus defining this group is critical. Clinician assessment prior to switching device is crucial, and there may be a role for more objective assessments. Commercially available devices such as the In-Check Dial G16 (Clement Clarke International) measure PIF and can vary internal resistance to mimic various devices. Device-specific products are also available, and can also be used to assess and train children in using a DPI [28]. Given the high prevalence of asthma, it may not be feasible to have such devices available for all paediatric asthma consultations; therefore, the development and validation of pragmatic assessment methods in children [69] would be advantageous. Furthermore, PIF is likely to be reduced during periods of acute illness, with one study reporting a drop of 11.0 L·min−1 [46]. The generation of a sufficient PIF to ensure adequate use of a DPI may be more challenging at times of asthma exacerbation. Other trials have examined DPI use during acute asthma, such as Drblik et al. [70], who included children who had already been assessed as being capable of using DPIs (and therefore did not meet criteria for inclusion in our analysis). They found no significant difference in FEV1, heart rate, oxygen saturation measured by pulse oximetry or tremor when comparing children treated with terbutaline via Turbuhaler with children treated with terbutaline via MDI in moderate to severe asthma. The majority of trials included in this systematic review assessed children with stable asthma, and excluded those with recent or current exacerbation. Only five studies assessed children with acute wheeze or asthma [32, 41, 46, 59, 60]. Considerable proportions of children aged <6 years (60% and 100%) in the studies assessing PIF did not meet study-defined criteria for adequate response [32, 41], and we found underpowered studies (n=22 and n=54) comparing DPI with MDI for spirometry or clinical response (respectively) [59, 60], with all studies excluding children presenting with severe exacerbations. Furthermore, none of the included studies compared patients during an acute exacerbation prior to the use of oral corticosteroids, or during moderate or severe exacerbations. The very low quality of the evidence makes it difficult to make robust recommendations for the use of DPIs in acute asthma, particularly for children aged <6 years; further research is required in this population. This is especially relevant to the design of trials of anti-inflammatory reliever approaches in the primary school-age group. Repeated education and review in asthma is a mainstay of asthma management. Specific education programmes varied, but improved education was consistently reported to improve technique in both the PIF and the physician-assessment groups [43, 44, 49, 51, 56, 57]. Kamps et al. [49] found that 79% of pharmacy-educated (average 30 min) inhaler users carried out all steps correctly, compared with 39% of general practitioner-instructed (average 5 min) users. 100% of DPI users were able to demonstrate correct use following standardised instruction [49]. Technique improved with supervision and caregiver support, as well as with practice over time. This was particularly the case for children who had regular checks of technique and further targeted education as needed [45]. Agertoft et al. [36] compared the PIF of 198 children with asthma to 256 non-asthmatic children (all aged 3–15 years); the asthmatic children produced significantly higher PIF across all age groups. This may suggest that PIF improves with repeat exposure to and use of DPI devices, further underlining the importance of training, education and reassessment. Regular assessment of technique and targeted education is the responsibility of all healthcare providers caring for children with asthma, including doctors in both community and hospital settings, pharmacists dispensing devices and specialist nurses. Multiple DPIs are available commercially, and this was reflected in the wide range of devices studied. In general, newer devices are designed with ease of use in mind, and steps such as loading capsules or holding upright to avoid spillage are less relevant. Younger children may have difficulties with some steps (e.g. holding correctly or removing the cap) that can be overcome with carer supervision and assistance. It was also noted that some older studies used DPIs as the gold standard and compared them to MDIs (without a spacer); these were excluded as this is no longer considered the standard of care. There are limitations of this review that must be acknowledged, primarily related to the available evidence. Few studies evaluated children with acute asthma symptoms, which limits any conclusions regarding the use of DPIs in this scenario. Furthermore, the heterogeneity of devices used, and measurements used, limits the comparability of studies. Lastly, there were relatively few head-to-head comparisons of DPI devices to inform which is the optimal device for use in primary school-aged children. In conclusion, the majority of primary school-aged children have the ability to use a DPI with adequate training, support and practice. Therefore, the authors conclude that it is reasonable that DPIs are used in primary school-aged children with stable asthma. Importantly, not all children (especially younger ones) can use DPIs, so MDIs with spacers must remain available and clinicians should make a case-by-case decision regarding which device would be most appropriate for each patient, in consultation with the child and family. Resources that may assist clinicians and families with making this decision include videos demonstrating correct technique, standardised checklists for assessment, placebo DPI devices and PIF measurement devices. However, the authors consider that PIF measurement is not necessary to commence a DPI if adequate technique can be demonstrated. Consistent correct use and adherence remain challenges, but these are also issues with MDIs and spacers and do not appear to be significantly worse with DPIs. Ongoing regular review and re-education involving doctors and community services such as asthma nurses and pharmacists are key, including the use of device-specific action plans. Evidence regarding the use of DPIs during acute exacerbations, in acute care settings, is scarce. Therefore, children should be managed with an MDI and spacer for the delivery of a bronchodilator according to current protocols. Depending on the severity of exacerbation, children may be able to continue maintenance therapy via a DPI. While this review provides the evidence to inform the current use of DPIs in primary school-aged children with asthma, there are significant evidence gaps that need to be addressed, including how best to assess a child's ability to use a DPI and the suitability of DPI use in acute exacerbations. Supplementary material Please note: supplementary material is not edited by the Editorial Office, and is uploaded as it has been supplied by the author. Supplementary material 00455-2024.SUPPLEMENT