Estimating Use of Short-term Asthma Reliever Inhalers from

Electronic Prescription Records

Holly Tibble

1,2 a

, Aziz Sheikh

1,2,3 b

and Athanasios Tsanas

1,2,4

Usher Institute, University of Edinburgh, Edinburgh, U.K.

Asthma UK Centre for Applied Research, Edinburgh, U.K.

Health Data Research UK BREATHE Hub for Respiratory Health, University of Edinburgh, Edinburgh, U.K.

The Alan Turing Institute, London, U.K.

Keywords: Asthma, Inhaler, Reliever, Salbutamol, Electronic Health Records, Prescriptions.

Abstract: Asthma is a common chronic lung disease which can be effectively managed for most people through regular

use of inhaled controller therapy. Short-acting Beta-2 Agonists (symptom relievers; SABA) may also be

prescribed to be used as needed, however over-reliance may indicate poor symptom control. SABA usage can

be estimated from refill rate observed in prescribing records. This study was a secondary analysis of a Scottish

longitudinal dataset of linked primary and secondary care data. The aims of this study were to estimate the

mean inhaled SABA dose per day for people diagnosed with asthma in a large EHR database, and to examine

variation by demographic factors such as age, sex, and social deprivation. The prescriptions dataset contained

over 40 million prescriptions between 2009 and 2017. 1,987,119 asthma reliever prescription records were

identified (5% of all prescriptions), of which 97% were inhaled formulations. The Spearman correlation

coefficient between subsequent years of aggregated (median) daily estimated SABA from one person-year to

the next was 0.67. Higher median daily inhaled SABA amounts were statistically significantly associated

(Wilcoxon Rank-Sum test p-value<0.05) with being older, male, living in an area of higher deprivation, and

any non-inhaled SABA prescription.

1 INTRODUCTION

Asthma is a common chronic lung disease

characterised by inflammation of the airways and

hyper-responsiveness (sensitivity of the nerve

endings in the airways so they become easily

irritated) to stimuli including allergens, exercise, and

infections (World Health Organization, 2020).

Inflammation results in obstruction of the airways,

and can present as wheezing, chest tightness,

coughing and shortness of breath (American

Academy of Allergy Asthma & Immunology, 2020).

Asthma can be effectively managed for most

people through regular use of Inhaled Cortico-

Steroids (ICS) (Barnes, 1998; Barnes & Pedersen,

1993; Suissa, Ernst, Benayoun, Baltzan, & Cai,

2000), although additional therapies such as Long-

Acting Muscarinic Antagonists (LAMA) and

Monoclonal Antibody therapy (MAb) (British

https://orcid.org/0000-0001-7169-4087

https://orcid.org/0000-0001-7022-3056

https://orcid.org/0000-0002-0994-8100

Thoracic Society & SIGN, 2019; Global Initiative for

Asthma, 2019; Peters, Ferguson, Deniz, & Reisner,

2006) may be used in parallel for those with

insufficient control of their symptoms. Additionally,

most people are prescribed Short-Acting β-2 Agonist

(SABA) symptom reliever inhalers to be used as

needed (British Thoracic Society & SIGN, 2019;

Global Initiative for Asthma, 2019). SABA is a

bronchodilator (it opens the airways) which acts by

relaxing the muscles in the airways, with effects

lasting for around three to six hours (Ullman &

Svedmyr, 1988).

SABA may also be prescribed at higher doses in

non-inhaled formulations, including tablet, syrup,

injections or using a nebuliser (Pharmaceutical Press

Joint Formulary Committee, 2019a). Other reliever

medications are also available, such as

anticholinergics or theophylline (British Thoracic

Society & SIGN, 2019; Global Initiative for Asthma,

Tibble, H., Sheikh, A. and Tsanas, A.

Estimating Use of Short-term Asthma Reliever Inhalers from Electronic Prescription Records.

DOI: 10.5220/0010954700003123

In Proceedings of the 15th International Joint Conference on Biomedical Engineering Systems and Technologies (BIOSTEC 2022) - Volume 4: BIOSIGNALS, pages 311-318

ISBN: 978-989-758-552-4; ISSN: 2184-4305

311

2019; Peters et al., 2006). These may be used in

conjunction with controller medications, such as

inhalers containing both an ICS and a Long-Acting β-

2 Agonist (LABA), components (O’Byrne et al.,

2018), as alternatives to SABA, or simply as

additional reliever therapy.

While use of SABA inhalers is encouraged to

relieve exacerbations, both mild and severe, regular

usage is an indicator that current therapy is not

sufficient to achieve symptom control. The 2021

Global INitiative for Asthma (GINA) guidelines

classify SABA over-use as three or greater 200-dose

SABA inhalers within a year (Reddel et al., 2021).

Estimating an individual’s SABA usage may flag

when a step-up in therapy may be required, or when

an individual is not sufficiently adherent to their

routine controller therapy (Chan et al., 2020). This

can be used identify those currently at higher risk of

adverse outcomes, including the need for

unscheduled care, morbidity, and mortality

(Canonica et al., 2021; Stanford, Shah, D’Souza,

Dhamane, & Schatz, 2012).

Electronic Health Records (EHRs) can be used in

pragmatic observational and intervention studies of

asthma (Varsano, Segev, & Shitrit, 2017). SABA

usage can be estimated from prescribing records by

recording the rate of SABA cannister refill. While

previous studies have reported reliever use estimated

from prescribing records, the methodology has not

been sufficiently described for reproducibility or

validated. For example, many studies measure SABA

use averaged over a year (Blakey et al., 2017;

Disantostefano, Boudiaf, Stempel, Barnes, &

Greening, 2016; FitzGerald, Tavakoli, Lynd, Al

Efraij, & Sadatsafavi, 2017; Lugogo et al., 2021;

Makhinova, Barner, Richards, & Rascati, 2015;

Stanford et al., 2012), however this leads to

substantial delay in the detection of increased use.

The aims of this paper were to: (1) estimate the

mean inhaled SABA dose per day for people

diagnosed with asthma in a large EHR database, and

(2) examine the trends of mean inhaled SABA dose

with regards to consistency over time and variation

by demographic factors such as age, sex, and social

deprivation.

2 METHODS

2.1 Data

The Asthma Learning Healthcare System (ALHS)

dataset was created to develop and validate a

prototype learning healthcare system for asthma

patients in Scotland, in which patient data are used to

generate a continuous loop of knowledge-generation,

evidence based clinical practice change, and change

assessment/validation (Soyiri et al., 2018). Over half

a million patients from 75 general practices in

Scotland were recruited, with primary care records

linked to national accident and emergency, hospital

and mortality datasets using the Scottish health

identification number known as the Community

Health Index (CHI) (Tibble et al., 2019). The

prescriptions dataset contained 41,432,295 valid

prescription records for 671,298 individuals dated

between January 31

, 2009, to March 31

, 2017.

2.2 Reliever Medication Processing

To identify asthma medications, the medication’s

name was searched for “SALBUTAMOL” or any of

the brand names listed in Table 1 (extracted from the

British National Formulary Version 80; September

2020 (Pharmaceutical Press Joint Formulary

Committee, 2019a)).

Reliever medications with non-inhaled

formulations (tablets, injections, or nebulising

solution) were identified. For inhaled medications,

the mean inhaled SABA dosage per day was

estimated from the dosage, the quantity prescribed,

and the dates between prescriptions.

The medication strength (micrograms of

salbutamol in each dose unit, e.g., inhaler puff) was

then extracted by searching the medication name for

any of the values 95, 100, 200, or 400, followed by

“MICRO” or “MCG” (with or without a preceding

space). The volume of the inhaler was estimated by

searching the medication name for 60, 100, 120, or

200, following by “DOSE” (with or without a

preceding space or hyphen). The most common pack

sizes per brand and strength were then examined so

that the modal value could be imputed.

The 99.9

percentile of the dispensed quantity of

inhalers in a single prescription was calculated to be

used as the upper threshold for outliers, such as

recording the prescription of a 200-dose inhaler

mistakenly as the prescription of 200 inhalers. In such

cases, the recorded quantity was replaced with the

value 1 (a single inhaler).

Finally, multiple prescriptions of inhaled

salbutamol collected on the same day (multiple

prescriptions for single inhaler units rather than a

single prescription for multiple units) were condensed

into a single prescription.

SERPICO 2022 - Special Session on Diagnostic, Prognostic, and Phenotyping Models from Mined Administrative Healthcare Data

312

Table 1: Salbutamol brand names, formulations, and

strengths.

Brand

Names

Formulation

Medication

Strength

Generic

Tablet 2mg, 4mg

Oral Solution 2mg/5ml

Pressurised

Inhaler

100mcg

Inhalation

Powder

100mcg, 200mcg

Nebulising

Solution

2.5mg/2.5ml,

5mg/2.5ml

Salamol

Pressurised

Inhaler

100mcg

Nebulising

Solution

5mg/2.5ml

Ventolin

Infusion

Ampoules

5mg/5ml

Injection 500mcg/1ml

Oral Solution 2mg/5ml

Pressurised

Inhaler

100mcg

Inhalation

Powder

200mcg

Nebules 2.5mg, 5mg

Nebulising

Solution

5mg/1ml

Airomir

Pressurised

Inhaler

100mcg

Salbulin

Inhalation

Powder

100mcg

AirSalb

Pressurised

Inhaler

100mcg

Ventmax Capsule 4mg, 8mg

Asmasal Inhalation

Powder

95mcg

Pulvinal

Salbutamol

Inhalation

Powder

200mcg

2.3 Analysis Plan

The asthma reliever prescription record identification

and cleaning process was reported. The number of

inhaled SABA cannisters prescribed per person-year

(in which at least one was prescribed) and time

between prescriptions was summarised.

The mean SABA dosage daily was then estimated

as the prescribed quantity of inhalers multiplied by

the medication strength in micrograms and the

volume of the inhaler unit, and then divided by the

number of days until the next prescription.

The deviation in estimates of daily inhaled SABA

usage by person and by person-year were estimated,

and the Spearman correlation coefficient between an

individual’s average from one year to the next was

evaluated.

The differences in inhaled SABA use associated

with non-inhaled salbutamol use and demographic

factors (age, sex, and deprivation) were investigated

using Wilcoxon Rank Sum tests. For the non-binary

variables (age and deprivation), increments of ordinal

categories were assessed in a pairwise manner. The

Scottish Index of Multiple Deprivation (SIMD) is a

composite geographic-level measure incorporating

income, employment, education, health, access to

services, crime and housing (Scottish Government

National Statistics Publications, 2016), reported here

in quintiles.

3 RESULTS

3.1 Reliever Medication Identification

Of the 41,432,295 valid prescription records,

1,987,119 (4.8%) were identified as salbutamol

products. 68,265 (3.4%) of the prescriptions were

identified as relating to non-inhaled formulations.

For 65.4%, no brand name was listed, indicating

either that it was a generic, or that the recorded

information was incomplete. 27.1% of records were

for Ventolin, and the remaining 7.6% had other

recorded brands.

1,918,854 of the prescriptions related to inhaled

salbutamol, of which 79.4% were unbranded, 9.4%

were Ventolin, 9.2% were Salamol, and 2.6% were

Airomir (remaining 0.4% of prescriptions were for

other brands). The most commonly prescribed

strength was 100mcg (94.1% of prescriptions),

followed by 200mcg (5.4%), 95mcg (0.5%), and

400mcg (<0.1%). Asmasal was the most common

brand at 95mcg (44.5%), and 55.5% were generic.

100% of the 400mcg prescriptions were generic.

The number of doses per cannister was not

recorded in the prescription data for 84.7% of

prescriptions. Table 2 shows the most common

number of doses that could be extracted from each

brand and strength, the percentage of extractable

records with that value, and the number of records for

which no value was extractable.

Estimating Use of Short-term Asthma Reliever Inhalers from Electronic Prescription Records

313

Table 2: The most common doses per cannister extracted by brand name and strength from inhaled salbutamol prescriptions.

Brand

Strength

(micrograms)

Doses per

Cannister

Percentage of prescriptions

with this value extracted

Percentage of prescriptions

no value extractable

Generic /

Brand not

listed

95 200 100% 48.9%

100 200 100% 84.9%

200 60 64.8% 64.9%

400 200 100% 98.1%

Airomir 100 200 100% 77.7%

Asmasal 95 200 99.8% 81.5%

Pulvinal 200 100 98.0% 89.0%

Salamol 100 200 100% 84.4%

Salbulin 100 200 100% 96.9%

Ventolin

100 200 100% 90.6%

200 60 99.0% 89.8%

Consequently, Pulvinal was imputed as 100

doses, Ventolin and generic (or brand not listed)

200mcg as 60 doses, and all other brand-strengths as

200 doses. In total, 94.8% of prescriptions were

estimated to have 200 doses, 4.9% to have 60 doses,

and 0.3% to have 100 doses.

Less than 0.01% of records had missing quantity,

and 0.07% had quantity above the calculated outlier

threshold of 4 inhalers. After imputing these records

to have a single unit quantity, the quantity of cannisters

prescribed was estimated to be 1 for 50.1% of

prescriptions, 2 for 48.5%, 3 for 0.8%, and 4 for 0.6%.

Finally, multiple prescriptions of single inhaled

salbutamol cannisters collected on the same day were

condensed into a single prescription, resulting in

1,882,586 person-days.

3.2 Inhaled Salbutamol Use

In years containing at least one inhaled SABA

prescription, a median of 3 cannisters were obtained

(range 1 – 262, interquartiles 1-6). This was equal to

a median of 420 doses obtained each year (range 1-

52400, interquartiles 200-1200). Using the GINA

criteria (Reddel et al., 2021), a median of 66.7% of

each individual’s person-years could be classed as

periods of over-use (range 11.1% - 100%,

interquartiles 37.5% - 88.9%).

For each individual’s last prescription during the

study period (n=149,621), the duration that the

prescription lasted for could not be calculated (for

29.3% of individuals, this was their only prescription

in the study period). For the remaining prescriptions,

the median time until a repeat was filled was 48 days

(interquartiles 27 and 93 days), with a range of 1 to

2981 days.

The median estimated inhaled salbutamol amount

was 124.2mcg per day, with interquartiles of 45.2 and

360.4mcg. The range of observed daily estimates was

4.1 to 200,000mcg, with the latter being roughly

equivalent to 8 full inhalers being consumed on a

single day (and thus clearly erroneous). Outlier

values were observed when high amounts were

obtained and then a subsequent refill was made

shortly after. For example, there were 7618 (0.4%)

cases in which the estimated daily use was equal to

the amount dispensed, as another prescription was

obtained the very next day.

The InterQuartile Range (IQR) for the estimated

inhaled salbutamol amount was 315.2mcg. There

was less variation within each individual’s estimates,

with a median within-person IQR of 82.0. The

variation within a person was roughly equal to the

variation within a person-year, however, with a

median within-person-year IQR of 93.7. The

Spearman correlation for an individual’s median

estimate from one year to the next was 0.67.

The median daily inhaled salbutamol amount

increased across age categories: 77.5mcg in the under

18s, 88.9mcg in the 18-35s, 136.4mcg in the 36-55s,

180.2mcg in the 56-75s, and 202.0mcg in the those

aged over 75 (Wilcoxon Rank Sum test p<0.001 for

each age group increment’s pairwise difference). The

median amount also increased consistently from

highest SIMD quintile (least deprived) to lowest, at

97.3mcg, 106.4mcg, 121.6mcg, 141.8mcg, and

159.4mcg, respectively (p<0.001 for each quintile

increment’s pairwise difference). Finally, the median

daily amount was higher for men (135.1mcg) than

women (115.6mcg, p<0.001). This finding was

consistent across age groups, as shown in Figure 1.

SERPICO 2022 - Special Session on Diagnostic, Prognostic, and Phenotyping Models from Mined Administrative Healthcare Data

314

Table 3: Median and Interquartile Range of Daily Inhaled Salbutamol (micrograms) by demography.

Factor Level Median IQR

Sex

Male 135.14 48.54 - 387.10

Female 115.61 42.64 – 387.10

Age Group

Under 18 77.52 33.78 – 190.00

18-35 88.89 34.42 – 277.78

36-55 136.36 47.17 – 394.74

56-75 180.18 60.42 – 476.19

Over 75 202.02 73.26 – 487.80

Scottish Index of

Multiple Deprivation

1 (Most deprived) 159.36 52.22 – 444.44

2 141.84 48.43 – 408.16

3 121.58 45.05 – 357.14

4 106.38 41.11 – 312.50

5 (Least deprived) 97.32 39.45 – 270.27

Figure 1: Boxplot of the Daily Inhaled Salbutamol (micrograms), stratified by Age Group and Sex. The central dot denotes

the median value, and the error bars denote the interquartile range.

3.3 Non-inhaled Salbutamol Use

5799 people had at least one prescription for non-

inhaled salbutamol (tablets, injections, or nebulising

solution) during the study period, of whom 5058 also

had inhaled salbutamol (87.2%). Conversely, only

3.4% of individuals with prescriptions for inhaled

salbutamol also had non-inhaled salbutamol

prescriptions. Those with at least one prescription for

non-inhaled salbutamol had a median of three such

prescribing events, with interquartiles 1 to 13 (range

1 to 254).

The median daily inhaled salbutamol amount was

lower (120.5mcg, interquartiles 44.3 to 354.0mcg) in

person-years without non-inhaled salbutamol

prescribed, compared to years in which at least one

prescription for non-inhaled salbutamol was filled

(496.9mcg, interquartiles 200.0 to 975.6mcg,

Wilcoxon Rank Sum test p<0.001). Indeed, it was

also higher in those with any non-inhaled salbutamol

prescribed during the study period (384.6mcg,

interquartiles 143.4 to 800.0mcg) compared to those

with none (117.0mcg, interquartiles 43.3 to

344.8mcg, p<0.001).

4 DISCUSSION

4.1 Principal Findings

5% of prescriptions were identified as salbutamol-

based asthma symptom reliever medications, of

which 97% were inhaled formulations. In years

containing at least one inhaled SABA prescription, a

Estimating Use of Short-term Asthma Reliever Inhalers from Electronic Prescription Records

315

median of 3 cannisters were obtained. Using the

GINA threshold of 600-doses (Reddel et al., 2021), a

median of 66.7% of each individual’s person-years

could be classed as periods of over-use (range 11.1%

- 100%, interquartiles 37.5% - 88.9%). The median

time until a repeat was filled, for people with more

than one SABA prescription, was 48 days.

The median estimated daily inhaled salbutamol

amount was 124.2mcg (interquartiles 45.2 and

360.4mcg). The median of the interquartile range for

daily estimates within a person-year was 82.0mcg,

and there was moderate-strong correlation between

the median from one person-year to the next

(Spearman correlation coefficient = 0.67).

In a series of pairwise Wilcoxon Rank Sum tests,

higher median daily inhaled salbutamol was found to

be significantly associated with being older, male, or

living in an area of higher deprivation. The median

daily inhaled salbutamol amount was higher in the

3.4% of individuals with any non-inhaled salbutamol

prescribed during the study period. For these

individuals, the median daily inhaled salbutamol was

higher in years in which non-inhaled salbutamol was

also prescribed.

4.2 Results in Context

Many previous studies in the literature estimated

daily inhaled SABA use over the duration of a year

(Blakey et al., 2017; Disantostefano et al., 2016;

FitzGerald et al., 2017; Lugogo et al., 2021;

Makhinova et al., 2015; Stanford et al., 2012). In this

study, we found that substantial variation within a

year was commonplace, with a median within-

person-year interquartile range of 89.3mcg. Studies

such as Stanford et al. (2012), Blakey et al. (2017) ,

and Fitzgerald et al. (2017) have successfully

demonstrated that SABA use in the previous year is a

predictive factor for incidence of asthma attacks.

Assuming that SABA over-reliance is a contributory

factor to this increased risk, rather than simply acting

as a marker for increased asthma symptoms or

reduced symptom control, the ability to detect short-

term changes in SABA use may facilitate more timely

detection of increased risk in order to aid intervention.

Another common way of utilising asthma

symptom reliever inhalers for assessing symptom

control is to calculate the ratio of symptom controller

to reliever medication prescriptions. This may be

either as the number of prescriptions or cannisters

themselves (Baltrus et al., 2017; Lieu, Capra,

Quesenberry, Mendoza, & Mazar, 1999) or the ratio

of inhaled corticosteroid (typically converted to

budesonide equivalent as some medications such as

Ciclesonide are double potency) to salbutamol in

micrograms (Gonem, Cumella, & Richardson, 2019).

Finally, this study only used inhaled salbutamol

for the primary analyses, with prescriptions of other

formulations (including solutions for nebuliser

devices) evaluated as a potential risk modifier. Other

formulations may be prescribed for very young

children or those for whom inhaled therapy cannot be

used reliably, or in the case of acute exacerbation

(Pharmaceutical Press Joint Formulary Committee,

2019b). Other studies, such as Stanford et al.

converted nebulised doses to the equivalent inhaled

doses (Stanford et al., 2012), however the use of non-

inhaled formulations has been identified previously

as a strong risk predictor for adverse outcomes. For

example, Paris et al. found a 21.6 times increase in

hazard ratio for asthma-related hospitalisation for

every daily nebulised salbutamol use (Paris et al.,

2008).

One limitation of the assessment of inhaled

salbutamol use from prescription records is that

prescription refills do not perfectly correspond to

medication usage. This is a limitation of this type of

analysis for any type of medication, however for

reliever medications it may be more common. For

example, individuals may wish to have multiple

inhalers at different locations, and may also be more

likely to misplace inhalers due to their irregular use.

Estimating daily inhaled SABA use over multiple

refills may mitigate this affect somewhat, as periods

of overlap of inhalers may be evened out.

5 CONCLUSIONS

In this study, we have estimated the median inhaled

SABA dose per day and the proportion of individuals

with GINA-defined SABA over-use from our

Scottish cohort, and identified factors associated with

higher inhaled SABA use: increased age, male sex,

higher deprivation, and history of non-inhaled SABA

prescription.

REFERENCES

American Academy of Allergy Asthma & Immunology.

(2020). Asthma Defined. Retrieved November 25,

2020, from https://www.aaaai.org/conditions-and-

treatments/conditions-dictionary/asthma

Baltrus, P., Xu, J., Immergluck, L., Gaglioti, A., Adesokan,

A., & Rust, G. (2017). Individual and county level

predictors of asthma related emergency department

visits among children on Medicaid: A multilevel

SERPICO 2022 - Special Session on Diagnostic, Prognostic, and Phenotyping Models from Mined Administrative Healthcare Data

316

approach. Journal of Asthma, 54(1), 53–61.

https://doi.org/10.1080/02770903.2016.1196367

Barnes, P. J. (1998). Efficacy of inhaled corticosteroids in

asthma. Journal of Allergy and Clinical Immunology,

102(4), 531–538. https://doi.org/10.1016/S0091-

6749(98)70268-4

Barnes, P. J., & Pedersen, S. (1993). Efficacy and Safety of

Inhaled Corticosteroids in Asthma. American Review of

Respiratory Disease, 148(4_pt_2), S1–S26.

https://doi.org/10.1164/ajrccm/148.4_Pt_2.S1

Blakey, J. D., Price, D. B., Pizzichini, E., Popov, T. A.,

Dimitrov, B. D., Postma, D. S., Thomas, M. (2017).

Identifying Risk of Future Asthma Attacks Using UK

Medical Record Data: A Respiratory Effectiveness

Group Initiative. Journal of Allergy and Clinical

Immunology: In Practice, 5(4), 1015–1024.

https://doi.org/10.1016/j.jaip.2016.11.007

British Thoracic Society, & SIGN. (2019). British guideline

on the management of asthma (2019 Edition).

Canonica, G. W., Paggiaro, P., Blasi, F., Musarra, A.,

Richeldi, L., Rossi, A., & Papi, A. (2021). Manifesto on

the overuse of SABA in the management of asthma:

new approaches and new strategies. Therapeutic

Advances in Respiratory Disease, 15,

175346662110425. https://doi.org/10.1177/175346662

11042534

Chan, A. H. Y., Katzer, C. B., Horne, R., Haughney, J.,

Correia de Sousa, J., Williams, S., & Kaplan, A. (2020).

SABA Reliance Questionnaire (SRQ): Identifying

Patient Beliefs Underpinning Reliever Overreliance in

Asthma. Journal of Allergy and Clinical Immunology:

In Practice, 8(10), 3482-3489.e1. https://doi.org/

10.1016/j.jaip.2020.07.014

Disantostefano, R. L., Boudiaf, N., Stempel, D. A., Barnes,

N. C., & Greening, A. P. (2016). The frequency of, and

adherence to, single maintenance and reliever therapy

instructions in asthma: A descriptive analysis. Npj

Primary Care Respiratory Medicine, 26(May), 1–5.

https://doi.org/10.1038/npjpcrm.2016.38

FitzGerald, J. M., Tavakoli, H., Lynd, L. D., Al Efraij, K.,

& Sadatsafavi, M. (2017). The impact of inappropriate

use of short acting beta agonists in asthma. Respiratory

Medicine, 131, 135–140. https://doi.org/10.1016/

j.rmed.2017.08.014

Global Initiative for Asthma. (2019). Pocket Guide for

Asthma Management and Prevention 2019. Retrieved

from www.ginasthma.org.

Gonem, S., Cumella, A., & Richardson, M. (2019). Asthma

admission rates and patterns of salbutamol and inhaled

corticosteroid prescribing in England from 2013 to

2017. Thorax, 74, 705–706. https://doi.org/10.1136/

thoraxjnl-2018-212723

Lieu, T. A., Capra, A. M., Quesenberry, C. P., Mendoza, G.

R., & Mazar, M. (1999). Computer-based models to

identify high-risk adults with asthma: Is the glass half

empty or half full? Journal of Asthma,

36(4), 359–370.

https://doi.org/10.3109/02770909909068229

Lugogo, N., Gilbert, I., Tkacz, J., Gandhi, H., Goshi, N., &

Lanz, M. J. (2021). Real-world patterns and

implications of short-acting β2-agonist use in patients

with asthma in the United States. Annals of Allergy,

Asthma and Immunology, 126(6), 681-689.e1.

https://doi.org/10.1016/j.anai.2021.01.024

Makhinova, T., Barner, J. C., Richards, K. M., & Rascati,

K. L. (2015). Asthma Controller Medication

Adherence, Risk of Exacerbation, and Use of Rescue

Agents Among Texas Medicaid Patients with Persistent

Asthma. Journal of Managed Care & Specialty

Pharmacy, 21(12), 1124–1132. https://doi.org/

10.18553/jmcp.2015.21.12.1124

O’Byrne, P. M., FitzGerald, J. M., Bateman, E. D., Barnes,

P. J., Zhong, N., Keen, C., Reddel, H. K. (2018).

Inhaled Combined Budesonide–Formoterol as Needed

in Mild Asthma. New England Journal of Medicine,

378(20), 1865–1876. https://doi.org/10.1056/NEJMo

a1715274

Paris, J., Peterson, E. L., Wells, K., Pladevall, M.,

Burchard, E. G., Choudhry, S., Williams, L. K. (2008).

Relationship between recent short-acting β-agonist use

and subsequent asthma exacerbations. Ann Allergy

Asthma Immunol., 101(5), 482–487. Retrieved from

https://www.ncbi.nlm.nih.gov/pmc/articles/

PMC3624763/pdf/nihms412728.pdf

Peters, S. P., Ferguson, G., Deniz, Y., & Reisner, C. (2006).

Uncontrolled asthma: A review of the prevalence,

disease burden and options for treatment. Respiratory

Medicine, 100, 1139–1151. https://doi.org/10.1016/

j.rmed.2006.03.031

Pharmaceutical Press Joint Formulary Committee. (2019a).

British National Formulary Version 80. (BMJ Group

and Pharmaceutical Press, Ed.) (80th ed.). London.

Pharmaceutical Press Joint Formulary Committee. (2019b).

Salbutamol. In British National Formulary Version 80.

Reddel, H. K., Bacharier, L. B., Bateman, E. D., Brightling,

C. E., Brusselle, G. G., Buhl, R., Boulet, L.-P. (2021).

Global Initiative for Asthma (GINA) Strategy 2021 –

Executive Summary and Rationale for Key Changes.

American Thoracic Society. https://doi.org/

10.1164/rccm.202109-2205pp

Scottish Government National Statistics Publications.

(2016). Introducing The Scottish Index of Multiple

Deprivation 2016. Retrieved from https://www.gov.

scot/Resource/0050/00504809.pdf

Soyiri, I. N., Sheikh, A., Reis, S., Kavanagh, K., Vieno, M.,

Clemens, T., Simpson, C. R. (2018). Improving

predictive asthma algorithms with modelled

environment data for Scotland: an observational cohort

study protocol. BMJ Open, 8, e23289. https://doi.org/

10.1136/bmjopen-2018-023289

Stanford, R. H., Shah, M. B., D’Souza, A. O., Dhamane, A.

D., & Schatz, M. (2012). Short-acting β-agonist use and

its ability to predict future asthma-related outcomes.

Annals of Allergy, Asthma and Immunology, 109(6),

403–407. https://doi.org/10.1016/j.anai.2012.08.014

Suissa, S., Ernst, P., Benayoun, S., Baltzan, M., & Cai, B.

(2000). Low-Dose Inhaled Corticosteroids and the

Prevention of Death from Asthma. New England

Journal of Medicine, 343(5), 332–336. https://doi.org/

10.1056/NEJM200008033430504

Estimating Use of Short-term Asthma Reliever Inhalers from Electronic Prescription Records

317

Tibble, H., Tsanas, A., Horne, E., Horne, R., Mizani, M.,

Simpson, C. R., & Sheikh, A. (2019). Predicting asthma

attacks in primary care: protocol for developing a

machine learning-based prediction model. BMJ Open,

9(7), e028375. https://doi.org/10.1136/BMJOPEN-

2018-028375

Ullman, A., & Svedmyr, N. (1988). Salmeterol, a new long

acting inhaled Beta-2 adrenoceptor agonist: comparison

with salbutamol in adult asthmatic patients. Thorax,

43(May), 674–678.

Varsano, S., Segev, D., & Shitrit, D. (2017). Severe and

non-severe asthma in the community: A large electronic

database analysis. Respiratory Medicine, 123, 131–139.

https://doi.org/10.1016/j.rmed.2016.12.017

World Health Organization. (2020). WHO | Asthma:

Definition.

SERPICO 2022 - Special Session on Diagnostic, Prognostic, and Phenotyping Models from Mined Administrative Healthcare Data

318