View metadata, citation and similar papers at core.ac.uk brought to you by CORE

provided by Elsevier - Publisher Connector Respiratory Medicine (2012) 106, 1743e1748

Available online at www.sciencedirect.com

journal homepage: www.elsevier.com/locate/rmed

Lack of spirometry use in Danish patients initiating targeting

Mette Marie Koefoed*, Rene´ dePont Christensen, Jens Søndergaard, Dorte Ejg Jarbøl

Research Unit of General Practice, Institute of Public Health, University of Southern Denmark, J.B. Winsløws Vej 9A, 1, DK-5000 Odense C, Denmark

Received 23 March 2012; accepted 17 September 2012 Available online 5 October 2012

KEYWORDS Summary Obstructive lung Background: Research indicates that a large proportion of patients using medication targeting disease; obstructive lung disease have no history of spirometry testing. Spirometry; Objective: To investigate the use of spirometry when initiating pulmonary medication target- Medication ing obstructive lung disease and to explore possible patient characteristics associated with undergoing spirometry. Methods: Population-based cohort study. Three Danish National registers were linked enabling a retrieval of data on all primary and secondary healthcare services provided in the time period 2007e2010. Results: In 2008 a total of 40,969 patients were registered as first time users of pulmonary medication targeting obstructive lung disease. The mean age of the study cohort was 55.6 yrs (SD 18.7). Spirometry test had been performed in 20,262 (49.5%) of the study cohort in the period from 6 months before to 12 months after their first prescription. Just above one third of the cohort, 14,275 (34.8%), had undergone spirometry in the two-month period close to redemption of their first prescription. Women and patients in the oldest age categories were less likely to have spirometry performed. Conclusions: Many patients initiate medication targeting obstructive pulmonary disease without having airway obstruction confirmed through spirometry. Only one third of the study cohort had a spirometry performed when initiating medication and half had still not undergone spirometry after a year. There should be an increased focus on confirming airway obstruction when initiating medication. ª 2012 Elsevier Ltd. All rights reserved.

* Corresponding author. Tel.: þ45 6550 4244. E-mail address: [email protected] (M.M. Koefoed).

0954-6111/$ - see front matter ª 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.rmed.2012.09.012 1744 M.M. Koefoed et al.

Introduction The Danish National Prescription Register records all prescribed that are redeemed. Each prescription Spirometry is recommended as the gold standard for con- record includes the patient’s CPR, identification code of the firming obstructive lung disease and is a cornerstone in diag- prescriber, date of purchase, the brand name, anatomical nosing the diseases and chronic obstructive pulmonary therapeutic chemical system code (ATC), dose unit and disease (COPD).1,2 Initiating medication for obstructive lung quantity. The daily dose and indication are not recorded in the 22 disease without spirometry entails the risk that patients use database. All medication targeting obstructive lung disease, this medication without evidence of airway obstruction. defined as all medications within the ATC code R03, requires Neither clinical examination nor history taking can diagnose a prescription and registration is therefore complete. Medi- COPD adequately or distinguish it from asthma or other cation with ATC code R03 include: Beta-2-agonists, anticho- illnesses also giving respiratory symptoms.3e5 Although linergic drugs, inhaled corticosteroids, , medication relieves symptoms, reduces risk of exacerbations , sodium cromoglycate, and . and improves quality of life,6,7 it is not without risk8,9 and spirometry should be performed to ensure correct diagnosis.10 Sampling algorithm: identification of first time users However, studies have shown that a large proportion of of medication targeting obstructive lung disease patients diagnosed with COPD or asthma have no history of 11e14 spirometry testing. Further, it has been indicated that The Danish National Prescription Register was used to iden- many patients are prescribed medication for obstructive lung tify all adults who were first time users of medication tar- disease without a relevant diagnosis of COPD or asthma 15,16 geting obstructive lung disease in 2008. Selection criteria registered. We found no studies estimating the use of were as follows: 1) All patients redeeming drugs with ATC spirometry when initiating medication targeting obstructive code R03 in 2008 were identified, 2) patients under 18 years lung disease. Some studies have shown gender differences in of age on 1 January 2008 were excluded, and 3) patients with disease management17 and age has also been identified as 18 records of previously redeemed prescriptions with ATC code a factor influencing whether spirometry is performed. The R03 in the prescription database (1995e2007) were aim of this study was therefore to assess to what extent excluded. Sampling algorithm is demonstrated in Fig. 1.The spirometry is conducted when initiating medication targeting date the first prescription was redeemed was defined as the obstructive lung disease, and to explore possible patient index date. Patients were defined as having repeated characteristics associated with undergoing spirometry. redemption of medication for obstructive lung disease if they redeemed more than one prescription of medication for Methods obstructive lung disease (ATC R03) within a one-year interval and the interval between the two prescriptions exceeded 30 days. Number of therapies redeemed in the first year was This population-based cohort study links three national assessed for each patient within the three main categories of registers: the Danish National Prescription Register, the R03 medication: Beta-2-agonists, and Danish National Patient Register and the Danish National inhaled corticosteroids. Other medications within the ATC Health Service Register. These registries cover the entire R03 category were rarely prescribed and were therefore population of Denmark and contain information on all excluded from analyses. healthcare services provided. All Danish citizens are assigned a unique civil registration number (CPR), which enables accurate linkage between these registries.19 Spirometry Denmark has a tax-funded public healthcare system, providing free access to general practice and hospital care. Records of spirometric procedures provided in the time More than 98% of Danish citizens are registered with a GP period 2007e2010 were extracted from the Danish National who act as a gatekeeper to the rest of the healthcare system by carrying out initial diagnostic investigations and referring patients to secondary care if necessary.

The Danish national registries

The Danish National Health Service Register contains information on all services provided in primary health care. Each registration includes the patients CPR, identification code for type of activity performed, e.g. spirometry, identification code for the primary healthcare clinic providing the health service and the week for reimburse- ment of the healthcare service.20 The Danish National Patient Register records all hospital admissions, contacts to emergency departments and contacts to outpatient clinics. Each record includes the patients CPR, the date of admission, data on the hospital and diagnostic and procedure codes including spirometry.21 Figure 1 Sampling algorithm. Lack of spirometry use in Danish patients 1745

Health Service Register and the Danish National Patient Table 1 Characteristics of first time users of medication Register. Patients were categorized as having the spirometry targeting obstructive lung disease. Total number 40,969. performed in general practice, other primary care clinics or hospital setting. For each patient we assessed if spirometry Characteristics N % was conducted in an 18-month period from 6 months before Gender to 12 months after the index date. The shortest time interval Male 19,083 46.6 between the index date and spirometry was extracted. Female 21,886 53.3 Further, we assessed the proportion of patients who had Age spirometry performed in a two-month period from one month Male 56.3 yrs before to one month after the index date. (SD 18.1) Female 55.0 yrs Statistics (SD 19.1) Repeated redemption Patient characteristics are reported using means and stan- Yes 15,279 37.3 dard deviations (SD) to describe continuous variables and No 25,690 62.7 percentages (%) to describe categorical variables. Logistic Number of pulmonary medication initiated regression models were used to calculate unadjusted and Monotherapy 24,760 60.4 adjusted odds ratios (ORs) with 95% CIs for the associations Only beta agonist 17,709 between patient characteristics and spirometry in the 18- Only anticholinergics 1359 month period. The patient characteristics included were: Only inhaled steroids 4764 gender, age, number of types of pulmonary medication Others 928 initiated 12 months preceding the index date and if Two therapies 13,757 33.6 patients had redeemed prescriptions repeatedly. P- Beta 2123 values < 0.05 were considered statistically significant. All agonist þ anticholinergics statistical analyses were carried out using STATA 11 (STA- Beta 11,535 TACorp, College Station, TX, USA). agonist þ inhaled steroids Inhaled 99 steroids þ anticholinergics Ethics Three therapies Beta agonist þ anticholinergics 2452 6.0 This project is register-based and according to “The Act on þ inhaled steroids Research Ethics Review of Health Research Projects in Denmark” only questionnaire surveys and medical database research projects involving human biological material are required to be notified to the research ethics committee. after their first prescription is shown in Table 2. A total of The research ethics committee has, therefore, not been 20,262 (49.5%) of the study cohort had spirometry per- contacted. The study was approved by the Danish Data formed, whereas only 14,275 of the study cohort (34.8%) Protection Agency, J.nr. 2011-41-5798. had a spirometry performed within the two-month period (data not shown). Among patients on monotherapy only Results 37.3% had spirometry performed, this increased to 65.7% among patients using two therapies and to 81.2% among Cohort characteristics patients using three therapies. In patients with repeated redemption of medication for obstructive lung disease, A total of 40,969 patients >18 years were identified as first 69.9% had spirometry performed. time users of medication targeting obstructive lung disease in The time distribution from the index date to spirometry 2008 (Fig. 1). The mean age of the study cohort of first time is shown in Fig. 3. The figure also demonstrates in which users was 55.6 years (SD 18.7). There was a slight predomi- setting the spirometry test was performed; the large nance of women (53.3% vs. 46.6%), and they were slightly majority were conducted in primary care. younger (55.0 yrs (SD 19.1) vs. 56.3 yrs (SD 18.1)), (Table 1). The age distribution is shown in Fig. 2. A total of 24,760 (60.4%) Patient factors associated with spirometry used only one type of pulmonary medication within the first year. Just about one third of the cohort 13,757 (33.6%) used Table 3 shows crude and adjusted odds ratios for associa- two types of pulmonary medication. The remaining 2452 tions between spirometry testing and age, gender, number (6.0%) of the cohort used all three types of pulmonary medi- of pulmonary medication initiated and repeated redemp- cation within the first year. A total of 15,279 (37.3%) had tion. Patients in age categories 28e47 years and above 68 redeemed medication prescriptions repeatedly, Table 1. years had statistically significantly lower chance of having spirometry performed in the 18-month period. For age Spirometry category 78e87 years there was an adjusted odds ratio 0.51 (95% CI 0.46e0.56) compared with the youngest age cate- The proportion of patients having undergone at least one gory (18e27 years) and for age category þ88 years an spirometry in the period from 6 months before to 12 months adjusted OR of 0.15 (95% CI 0.12e0.18) for having 1746 M.M. Koefoed et al.

N 4500 4000 3500 3000 2500 2000 men 1500 women 1000 500 0 18-27 28-37 38-47 48-57 58-67 68-77 78-87 88- age category (years)

Figure 2 Age distribution of first time users of medication targeted obstructive lung disease. Total number of patients N: 40,969. spirometry performed. Women were less likely to have A major strength is the population-based design spirometry performed with an adjusted OR of 0.86 (95% CI reflecting an entire population’s drug and health care 0.82e0.90). There was an increasing chance of undergoing utilization. Prescription data are complete; all redeemed spirometry with increasing number of pulmonary medica- prescriptions are registered, including all medication for tion initiated within the first year and with repeated obstructive lung disease, thereby capturing all first time redemption. users of this medication. Furthermore, our study comprises the entire healthcare system; all spirometry measurements from primary and secondary health care are assessed. The Discussion study may have some methodological limitations though; the Danish National Prescription Registry contains only This study captures all first time users of obstructive redeemed prescriptions and it is therefore not possible to pulmonary medication in Denmark in 2008. Only one third identify patients who do not redeem prescribed medication of these patients had spirometry performed when initiating and they will be misclassified. However, we do not consider medication and approximately half did still not have this misclassification to be a major problem as primary non- spirometry performed 12 months after initiating medica- compliance is considered small.23 The registry also has no tion. Women and patients in the oldest age categories were data on the dose or indication for the drug, making it less likely to have spirometry performed. impossible for us to differentiate between different obstructive pulmonary diseases. As we aimed to assess Table 2 The proportion of patients having received at whether spirometry is performed to confirm obstruction least one spirometry in the period from 6 months before to when medication targeting obstructive lung diseases is 12 months after their first prescription according to patient prescribed, we did not find diagnosis important for this characteristics. study. However, although confirming obstruction through spirometry is mandatory in diagnosing COPD and recom- Characteristics Spirometry Total N mended in asthma, repeated peak expiratory flow recorded N (%) measurements (PEF) confirming variation of obstruction can All 20,262 (49.5) 40,969 be used as an alternative diagnostic strategy in young Gender patients where COPD is unlikely. In Denmark all GP’s have Male 9970 (52.2) 19,083 access to spirometry, making this preferred diagnostic Female 10,292 (47.0) 21,886 strategy easy to perform. Although it cannot be excluded Age categories that PEF measurements to a minor extent might have 18e27 2123 (47.6) 4461 substituted spirometry in the youngest age categories, PEF 28e37 2495 (41.1) 6072 cannot be applied to the vast majority of the cohort and 38e47 2847 (46.1) 6179 therefore does not significantly influence our findings. 48e57 3307 (51.3) 6441 Identification of spirometric procedures depends on the 58e67 4486 (56.7) 7907 comprehensiveness of the Danish National Health Service 68e77 3333 (57.6) 5791 Register and the Danish National Patient Register. There are 78e87 1527 (45.5) 3357 no studies assessing the consistency of coding of spiro- 88e 144 (18.9) 761 metric procedures in these two registries. An under- Repeated redemption reporting to these registers would lead to an Yes 10,674 (69.9) 15,279 underestimation of spirometry testing. Reporting of No 9588 (37.3) 25,690 spirometry in primary health care is assumed to be high, Number of pulmonary medication initiated because this is prerequisite for reimbursement. In Monotherapy 9236 (37.3) 24,760 secondary care coding is done routinely, it increases Two therapies 9036 (65.7) 13,757 payment and is therefore also considered consistent. Three therapies 1990 (81.2) 2452 Although caution must be taken when interpreting data, we do not consider underreporting to be a significant problem. Lack of spirometry use in Danish patients 1747

N first prescription/index date obstructive lung disease. Joish et al. demonstrated that 10,000 patients diagnosed with COPD were using medication before 9,000 this diagnosis was registered and the median time from 8,000 Other primary care clinics 15 7,000 prescription to diagnosis was approximately one year. These findings facilitate our hypothesis that patients use 6,000 Hospital setting 5,000 medication without a relevant diagnosis and spirometry 4,000 testing is underused among medication users. Many studies General 3,000 Practice have shown an underutilization of spirometry among patients 11e14,18,24 2,000 with a diagnosis of asthma or COPD. A Swedish 1,000 study conducted by Arne et al. demonstrated that spirometry - -6 -5 -4 -3 -2 -1 1 2 3 4 5 6 7 8 9 10 11 12 was only documented in 59% of the patients in connection Month the spiromtery was performed with a diagnosis of COPD.13 Our study confirms that this Figure 3 Time from the first prescription of medication underutilization of spirometry is also present among patients targeting obstructive lung disease to spirometry*. *The shortest initiating pulmonary medication targeting these illnesses. time interval between the index date and spirometry is Our study explored a possible association between illustrated. spirometry and the patient’s age and gender. We have not found any other studies comparing gender differences in spirometry utilization among pulmonary medication users. Three studies have compared gender differences in We were unable to find published studies assessing the spirometry utilization among COPD patients; Watson utilization of spirometry in patients initiating medication et al.17 found an underuse of spirometry testing among targeting obstructive lung disease. The PLATINO study, by women (OR of 0.84, 95% C.I. 0.72e0.98) in concordance Montes et al.,16 estimated the prevalence of respiratory with Arne et al.,13 Han et al.18 found the opposite. Although medication use and spirometry utilization among these our study is not readily comparable with these studies we medication users in five Latin American cities. Of the patients conclude that gender difference still persists. Patients in reporting use of any or corticosteroid in the our study had a lower OR of having spirometry performed in previous 12 months, only 37.5% reported a history of age categories 28e47 years and in patients >68 years. The spirometry testing. That study demonstrated a lower OR was lowest in age categories 78e87 years and above 88 spirometry testing among medication users than the one years, in concordance with Han et al.’s18 findings and it is found in our study. This difference could de due to study noteworthy that both studies found a clear underuse in the design and local differences. An interesting finding was that oldest patients. There is no obvious explanation, but 70% of the patients receiving and/or corti- a similar finding has been seen in management of ischemic costeroids had no airway obstruction when tested with heart disease.25 Correct diagnosis is essential in the oldest spirometry; emphasizing the importance of confirmation of age groups, as increasing comorbidity increases the risk of obstruction in patients using medication targeting misdiagnosis. We found that increasing number of therapies

Table 3 Logistic regression analysis to assess if there is an association between gender, age, number of pulmonary medication received, repeated redemption and spirometry. N Crude OR 95% CI Adjusted OR 95% CI Gender Male 19,083 1 ee Female 21,886 0.81* 0.78e0.84 0.86* 0.82e0.90 Age category 18e27 4461 1 ee 28e37 6072 0.77* 0.71e0.83 0.73* 0.67e0.79 38e47 6179 0.94 0.87e1.02 0.85* 0.79e0.92 48e57 6441 1.16* 1.08e1.25 0.96 0.88e1.03 58e67 7907 1.44* 1.34e1,55 1.02 0.94e1.11 68e77 5791 1.49* 1.38e1.62 0.90* 0.83e0.98 78e87 3357 0.92 0.84e1.01 0.51* 0.46e0.56 88e 761 0.26* 0.21e0.31 0.15* 0.12e0.18 N. of pulmonary therapies initiated 1 24,760 1 ee 2 13,757 3.22* 3.08e3.36 2.27* 2.16e2.38 3 2452 7.24* 6.52e8.04 3.93* 3.51e4.40 Repeated redemption No 25,690 e Yes 15,279 3.89* 3.73e4.06 2.65* 2.52e2.78

*P-value < 0.05. 1748 M.M. Koefoed et al. initiated and repeated redemption both significantly 6. Calverley PM, Anderson JA, Celli B, et al. and flu- increased the chance of undergoing spirometry. This asso- ticasone propionate and survival in chronic obstructive ciation is not surprising, as they could both be a proxy for pulmonary disease. N Engl J Med 2007;356(8):775e89. disease severity. Only one third of the patients on mono- 7. Barr RG, Bourbeau J, Camargo CA, Ram FS. Tiotropium for therapy had spirometry performed, this increased to just stable chronic obstructive pulmonary disease: a meta-analysis. Thorax 2006;61(10):854e62. two thirds in patients initiating two therapies, and even in 8. Michele TM, Pinheiro S, Iyasu S. The safety of tiotropium e the patients initiating three therapies there is still one out of FDA’s conclusions. N Engl J Med 2010;363(12):1097e9. five who did not have spirometry performed. 9. Calverley PM, Stockley RA, Seemungal TA, et al. Reported This study demonstrates that a high percentage of patients pneumonia in patients with COPD: findings from the INSPIRE treated with medication targeting obstructive pulmonary study. Chest 2011;139(3):505e12. diseases do not undergo spirometry. We found a clear under- 10. Averame G, Bonavia M, Ferri P, et al. Office spirometry can use of spirometry in the oldest age categories and among improve the diagnosis of obstructive airway disease in primary women. Why prescribing is often done without confirmation of care setting. Respir Med 2009;103(6):866e72. obstruction through spirometry testing remains unanswered, 11. Weidinger P, Nilsson JL, Lindblad U. Adherence to diagnostic but efforts to increase spirometry utilization must be made. guidelines and quality indicators in asthma and COPD in Swedish primary care. Pharmacoepidemiol Drug Saf 2009; 18(5):393e400. Acknowledgments 12. Buffels J, Degryse J, Heyrman J, Decramer M. Office spirometry significantly improves early detection of COPD in general practice: the DIDASCO study. Chest 2004;125(4): We wish to thank Secretary Lise Keller Stark for proof- 1394e9. reading the manuscript. 13. Arne M, Lisspers K, Stallberg B, et al. How often is diagnosis of COPD confirmed with spirometry? Respir Med 2010;104(4): Funding 550e6. 14. Lange P, Rasmussen FV, Borgeskov H, et al. The quality of COPD This project was funded by the Danish Graduate School care in general practice in Denmark: the KVASIMODO study. Network in Public Health Science (GRASPH). The funder has Prim Care Respir J 2007;16(3):174e81. not had any influence on the study design, on the collec- 15. Joish VN, Brady E, Stockdale W, Brixner DI, Dirani R. Evaluating diagnosis and treatment patterns of COPD in primary care. tion, analysis or interpretation of data, on the writing of Treat Respir Med 2006;5(4):283e93. the manuscript or on the decision to submit the manuscript. 16. Montes de OM, Talamo C, Perez-Padilla R, et al. Use of respi- ratory medication in five Latin American cities: the PLATINO study. Pulm Pharmacol Ther 2008;21(5):788e93. Conflict of interest statement 17. Watson L, Vestbo J, Postma DS, et al. Gender differences in the management and experience of chronic obstructive pulmonary The authors report no conflicts of interest. The authors alone disease. Respir Med 2004;98(12):1207e13. are responsible for the content and writing of the paper. 18. Han MK, Kim MG, Mardon R, et al. Spirometry utilization for COPD: how do we measure up? Chest 2007;132(2):403e9. 19. Frank L. Epidemiology. When an entire country is a cohort. References Science 2000;287(5462):2398e9. 20. Andersen JS, Olivarius NF, Krasnik A. The Danish national 1. Rabe KF, Hurd S, Anzueto A, et al. Global strategy for the health service register. Scand J Public Health 2011; diagnosis, management, and prevention of chronic obstructive 39(Suppl. 7):34e7. pulmonary disease: GOLD executive summary. Am J Respir Crit 21. Lynge E, Sandegaard JL, Rebolj M. The Danish national patient Care Med 2007;176(6):532e55. register. Scand J Public Health 2011;39(Suppl. 7):30e3. 2. Bateman ED, Hurd SS, Barnes PJ, et al. Global strategy for 22. Kildemoes HW, Sorensen HT, Hallas J. The Danish national asthma management and prevention: GINA executive prescription registry. Scand J Public Health 2011;39(Suppl. 7): summary. Eur Respir J 2008;31(1):143e78. 38e41. 3. Renwick DS, Connolly MJ. Do respiratory symptoms predict 23. Tobi H, van den Heuvel NJ, de Jong-van den Berg LT. Does chronic airflow obstruction and bronchial hyperresponsiveness uncollected medication reduce the validity of pharmacy in older adults? J Gerontol A Biol Sci Med Sci 1999;54(3):M136e9. dispensing data? Pharmacoepidemiol Drug Saf 2004;13(7): 4. Straus SE, McAlister FA, Sackett DL, Deeks JJ. Accuracy of 497e500. history, wheezing, and forced expiratory time in the diagnosis 24. Bourbeau J, Sebaldt RJ, Day A, et al. Practice patterns in the of chronic obstructive pulmonary disease. J Gen Intern Med management of chronic obstructive pulmonary disease in 2002;17(9):684e8. primary practice: the CAGE study. Can Respir J 2008;15(1): 5. Buffels J, Degryse J, Liistro G, Decramer M. Differential diag- 13e9. nosis in a primary care population with presumed airway 25. Bond M, Bowling A, McKee D, et al. Does ageism affect the obstruction: a real-life study. Respiration 2012;84(1):44e54. management of ischaemic heart disease? J Health Serv Res Epub 2011 Nov 15. Policy 2003;8(1):40e7.