- Email: [email protected]
Pharmacy-specific quality indicators for asthma therapy
Research in Social and Administrative Pharmacy 1 (2005) 430–445
Pharmacy-specific quality indicators for asthma therapyq Almut G. Winterstein, Ph.D.*, Abraham G. Hartzema, Pharm.D., M.S.P.H., Ph.D.1 College of Pharmacy, Department of Pharmacy Healthcare Administration, University of Florida, PO Box 100496, Gainesville, FL 32610-0496, USA
Abstract Background: Lack of appropriate drug therapy in asthma patients is a recognized quality problem leading to preventable emergency room visits, hospitalizations, or death. While indicators measuring pharmacotherapy quality on the level of prescribers and third party payers are widely used, no such indicators exist for pharmacies. Objective: This study aimed to (1) develop quality indicators for asthma care applicable to retail pharmacies, (2) estimate the prevalence of inappropriate asthma drug therapy, and (3) explore variation in the quality of care across pharmacies. Methods: We present a descriptive analysis of automated patient-specific dispensing data from February 1, 2000 to January 31, 2001 of beta-agonists and antiinflammatory agents with a Food and Drug Administration–approved indication for asthma. Two quality indicators of potentially inappropriate drug therapy were applied: (1) the proportion of patients who obtained more than a 360-day supply
q This work was supported by a grant from the International Federation of Pharmacy (FIP) Foundation. Results were presented in part at the International Federation of Pharmacy Annual Meeting in Vienna, September 2002 and The American Pharmacy Association Annual Meeting in New Orleans, March 2003. * Corresponding author. Tel.: C1 352 273 6258; fax: C1 352 273 6270. E-mail addresses: [email protected]fl.edu (A.G. Winterstein), [email protected]fl.edu (A.G. Hartzema). 1 Perry A. Foote Chair in Health Outcomes Research and Pharmacoeconomics, University of Florida, USA.
1551-7411/$ - see front matter Ó 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.sapharm.2005.07.001
A.G. Winterstein, A.G. Hartzema / Research in Social and Administrative Pharmacy 1 (2005) 430–445
431
(assuming maximum acceptable daily dose) of short-term beta-agonists (SABA) including all inhaler types, inhaler solutions, and syrups during the 12-month study and (2) the proportion of patients with 2 consecutive early refills of defined SABAs. Indicator values with 95% confidence intervals are reported for each pharmacy. Results: Pharmacies had an average of 328 (range 169-534) patients who received SABAs. An average of 11 patients per pharmacy (3.4% of all patients who received SABAs) met the indicator 1 definition. The second indicator identified 8.2% (27) patients per pharmacy as short-term SABA overusers (range 3.9-11.9%). Of these, 48% did not receive any anti-inflammatory agents during the time frame when SABA overuse occurred. Conclusion: Application of drug therapy quality indicators at the level of individual pharmacies using dispensing data is feasible and identifies opportunities for quality improvement. Indicator 2 is most appropriate for daily practice, because it allows for timely identification of potentially uncontrolled patients, and offers a balance between indicator sensitivity and positive predictive value. Ó 2005 Elsevier Inc. All rights reserved. Keywords: Asthma; Quality indicators; Community pharmacy
1. Introduction Research over the past years suggests that suboptimal drug therapy may be one of the most prevalent components of medical care that causes preventable patient injury.1 While most medication error studies have focused on the acute care environment,2 enough is known to conclude that inappropriate drug therapy in ambulatory care contributes significantly to preventable morbidity and mortality.3 Among the high-risk areas for preventable adverse drug events in ambulatory care is the management of asthma. Asthma is responsible for approximately 1.1% of all hospitalizations, 1.5 million emergency department visits, and more than 5500 deaths in the United States annually.4,5 The rate of age-adjusted mortality for asthma has increased steadily despite significant advances in the available pharmacotherapeutic treatment.6 Randomized controlled trials and observational studies demonstrate that the adequate use of inhaled corticosteroids controls asthma symptoms and reduces asthma-related hospital admissions and deaths.7-9 Overuse of betaagonist inhalers, on the other hand, indicates a greater risk for asthmarelated death and near death.10 These findings have directed current treatment guidelines, which emphasize the greater use of anti-inflammatory medications and the judicious use of beta-agonists for short-time symptom relief.11 Despite these well-grounded guidelines, pharmacotherapy in asthma patients is not optimal. Various studies have described quality deficits in asthma care, including lack of diagnosis and treatment, overuse of
432
A.G. Winterstein, A.G. Hartzema / Research in Social and Administrative Pharmacy 1 (2005) 430–445
beta-agonists, and underuse of inhaled corticosteroids.12,13 It is therefore not surprising that the set of patient safety indicators forwarded by the Agency for Healthcare Research and Quality includes the rate of asthmarelated hospital admissions as one of its core measure of health care quality in the United States.14 Quality deficits in asthma treatment as well as other areas in medical care are measured by quality indicators, which are used to assess and benchmark health care quality at the provider, payer, or regional level. Indicator results are summarized in quality scores or report cards and increasingly influence accreditation procedures, consumer choices, and provider reimbursement. Quality assessments by federal constituencies focus traditionally on the hospital and institutional sectors, but the imminent prescription drug benefit for Medicare beneficiaries is expected to add drug utilization evaluations in ambulatory care.15 While pharmacies have the legal mandate for patient counseling and prospective medication reviews, these services are generally excused from formal quality assessments (eg, by the Joint Commission on Accreditation of Healthcare Organizations [JCAHO]). For example, the National Heart, Lung, and Blood Institute (NHLBI) guidelines on ‘‘the role of the pharmacist in improving asthma care’’ define specific tasks a pharmacist should perform. These tasks include patient education, instruction about inhaler techniques and peak flow use, and monitoring medication use.16 However, no specific quality indicator exists that would allow pharmacists to evaluate the quality of care for asthma patients at the level of their individual pharmacy, even though these indicators are available and used at the level of third party payer or physician groups. The development of such quality indicators would allow pharmacists to (1) evaluate and compare the current level of pharmaceutical care provided across pharmacies, and (2) identify and monitor the progress of pharmacybased quality improvement initiatives targeted toward uncontrolled asthma patients.
2. Methods The objectives of this study were to (1) develop quality indicators for asthma care that are applicable to retail pharmacies, (2) estimate the prevalence of inappropriate drug therapy in asthma patients per pharmacy, and (3) explore variation in the quality of care across pharmacies. The developed indicators were expected to meet the following conditions of feasibility and validity.17 Data should be easy to collect and interpret, facilitating initiatives for continuous quality improvement and monitoring change over time. Further, the method selected for data collection should be reliable (replicable and standardized). The quality indicators should be valid in detecting substandard asthma pharmacotherapy, and sensitive to detect
A.G. Winterstein, A.G. Hartzema / Research in Social and Administrative Pharmacy 1 (2005) 430–445
433
variation, ie, to identify differences between pharmacies or to detect changes over time. In addition, the indicators should detect substandard therapy highly predictable of adverse drug events, ie, improvements in indicator rates should likely improve patient outcomes. Two types of quality indicators, outcome-based and process-based, were identified from the literature. Outcome-based indicators focused on health care utilization associated with adverse events (emergency room visits or hospital admissions) that suggest uncontrolled asthma.14,18 Outcome-based indicators were not explored for 3 reasons. First, while these indicators deliver important information about asthma-related outcomes, they represent only indirectly the care processes that were inappropriate and preceded these outcomes; second, patient outcome data on the level of community pharmacies are difficult to retrieve; and third, even though the incidences of ER visits and hospital admissions at the national level are sizeable, they are expected to translate only in very small numbers of affected patients within a single pharmacy. Two indicators reflecting the quality of pharmacotherapy and based on pharmacy dispensing data were defined in this study. Indicator definitions presented follow the Agency of Healthcare Research and Quality criteria. These criteria include documentation of the clinical rationale with evidence sources supporting the measures, the operational description of the denominator and numerator, data sources and data-ascertainment detail, and indicator type.19 Both indicators are based on a cohort study by Spitzer et al10 that demonstrated a 2-fold higher risk for fatal and near-fatal events in patients who use more than 1 albuterol inhaler per month. The current NHLBI guidelines have adopted these findings and expanded their definition of overuse by limiting all short-acting beta-agonist inhalers to 1 per month.11 The study database was created by retrospective extraction of medication records from 8 randomly selected pharmacies of a major pharmacy chain in North Carolina. Extracted database entries included all beta-agonists and anti-inflammatory agents, with a Food and Drug Administration–approved indication for asthma, available in the U.S. market and dispensed between February 1, 2000 and January 31, 2001. Medications were identified by an NDC code and additional full text search using all brand and generic names. Ipratropium bromide and its combinations with albuterol were excluded. Entries related to nasal sprays, creams, ointments, lotions, and ophthalmic solutions were deleted after initial data extraction. A list of the considered medications, beta-agonists and anti-inflammatory agents, is presented in Table 1. For each database entry the following information was available: a unique patient and pharmacy identifier, drug name, dispensed quantity, and dispensing date. The first indicator used a 1-year time window to define long-term overuse of short-term beta-agonists (SABAs). All patients with at least 1 SABA
434
A.G. Winterstein, A.G. Hartzema / Research in Social and Administrative Pharmacy 1 (2005) 430–445
Table 1 Dispensing frequencies of beta-agonists and anti-inflammatory agents in 8 pharmacies between Feb 2000 and Jan 2001 (pooled results) No. of database entries
%
No. of patients
Beta-agonists Terbutaline inhalers Metaproterenol syrupa Terbutaline tabletsa Albuterol inhalation capsules Metaproterenol solution for inhalationa Pirbuterol inhalers Albuterol tablets (including sustained release)a Albuterol syrup Metaproterenol inhalers Salmeterol inhalers and inhalation capsulesa Albuterol solution for inhalation Albuterol inhaler
0 1 10 14 14 247 329 585 693 989 1352 4627
Total
8861
Anti-inflammatory agents Dexamethasone elixir/solution for inhalation Flunisolide inhaler Zileuton tablets Cortisone/hydrocortisone tablets Nedocromil inhaler Dyphylline/Guaifenesin tablets Fludrocortisone tablets Methylprednisolone tablets Budesonide inhaler/solution for inhalation Beclomethason inhaler Cromoglycin inhaler/solution for inhalation Dexamethasone tablets Triamcinolone inhaler Zafirlukast tablets Prednisolone syrup/solution/tablets Theophylline tablets/elixir Montelukast tablets Fluticasone inhaler/rotadisk
8 9 11 21 30 49 101 103 112 199 235 260 271 285 754 1083 1192 1254
0.1 0.2 0.2 0.3 0.5 0.8 1.7 1.7 1.9 3.3 4.0 4.4 4.5 4.8 12.7 18.1 19.9 21.0
8 9 3 5 8 23 22 100 62 88 113 148 120 53 580 198 314 509
Total
5977
100.0
2363
a
0 0.0 0.1 0.2 0.2 2.8 3.7 6.6 8.0 11.1 15.2 52.1 100
0 1 6 4 3 110 103 446 176 301 620 1984 3754
% 0 0.1 0.2 0.1 0.1 2.9 2.7 11.9 4.7 8.0 16.5 52.8 100
0.3 0.4 0.1 0.2 0.3 1.0 0.9 4.2 2.6 3.7 4.8 6.3 5.1 2.2 24.5 8.4 13.3 21.5 100
Not included in the definition of SABA for quality indicator development.
dispensing event in the 1-year period were included in the analysis. Overusers were defined as patients receiving more than 360 days supply within 12 months. All SABA derivates and dosage forms available in the U.S. market except long-acting beta-agonists, sustained release forms (salmeterol and all albuterol and terbutaline tablets), and combination products (eg, CombiventÒ) were included in the analysis. For each dosage form the dispensed amount was converted to the equivalent amount of days
A.G. Winterstein, A.G. Hartzema / Research in Social and Administrative Pharmacy 1 (2005) 430–445
435
of supply assuming the patient required the highest acceptable dose. Each albuterol, metaproterenol, or pirbuterol inhaler was counted as a 30-day supply, which equals the maximum monthly amount of SABA that should be used. Albuterol nebulizer solution, syrup, and rotacaps were transformed into therapeutic equivalent amounts of the respective inhaler formulation. These dosage forms were included because they constituted more than 30% of all filled beta-agonist prescriptions, and were therefore important for detecting overuse. Metaproterenol syrup and nebulizer solution were excluded from analysis because of their small dispensing frequencies. Appendix 1 illustrates drugs and dosage forms included in the analysis and, where applicable, the transformation in inhaler equivalents. Whenever multiple SABA formulations were dispensed at the same day, the dispensed doses were collapsed in a single dispensing event and expressed as the ‘‘total amount of days supply obtained’’. For example, when a patient obtained at a single day 150 unit-doses albuterol nebulizer solution 0.083% (Z45 days supply) and 1 albuterol metered-dose inhaler (30 days supply), a total of 75 days supply was obtained at this particular dispensing encounter. For each patient, the total number of days supply obtained at each dispensing event was summarized for the 1-year study period. The second indicator was based on a shorter time window and expressed as consecutive short-term overuse of SABAs. The denominator included all patients with at least 1 SABA dispensing event. Overuse was defined as 2 consecutive early SABA refills, where early refill is defined as a negative difference between the actual and predicted refill period. Refill intervals were analyzed to facilitate recognition of episodic uncontrolled asthma (occurring over periods shorter than 1 year). The principles for calculating indicator values followed those for calculating medication possession ratios, where the theoretical number of days supply obtained (based on the maximum allowable dose and the total volume dispensed) is divided by the actual number of days until a refill occurs. For this study, we calculated medication possession differences, in which the number of days supply obtained at one particular dispensing event is subtracted from the number of days until a refill occurred. Early refill is defined as any negative difference. Because early refills might occur independent of asthma control (eg, patient happens to stop by the pharmacy or obtains larger supply due to travel plans), short-term SABA overuse was defined as the occurrence of two consecutive early refills. Appendix 2 illustrates the necessary computations and the indicator criteria. We further tested whether the identified SABA overusers received any anti-inflammatory agent during the defined period of overuse. The supporting clinical rationale was derived from several studies, demonstrating an association between anti-inflammatory therapy and reduced morbidity and mortality rates of asthma, as well as the current NHLBI treatment guidelines, recommending the use of anti-inflammatory agents as soon as regular SABA use occurs.7,8,20,21
436
A.G. Winterstein, A.G. Hartzema / Research in Social and Administrative Pharmacy 1 (2005) 430–445
Indicator values were calculated using MS Access 2000Ò and expressed in percent with 95% confidence intervals. Exact confidence intervals (binomial distribution) were calculated using the average sample size of all stores rather than store-specific sample sizes to account for the variation in sample size by store. 3. Results 3.1. Descriptive data analysis The database included 8861 prescriptions of beta-agonists dispensed to a total of 3754 patients (Table 1). More than three-quarters of all filled prescriptions were for albuterol. The number of prescriptions for betaagonists varied across pharmacies. In the 12-month period, between 504 and 1584 beta-agonist prescriptions were filled in the 8 study pharmacies, but the distribution of dispensing frequencies for the different asthma drugs and dosage forms were similar across pharmacies. The largest variation occurred in the proportion of entries for albuterol metered-dosed inhalers (MDI), which varied between 46.0% and 60.7% across pharmacies, and salmeterol inhalers, which varied between 6.4% and 17.4%. The 8 study pharmacies had 4627 albuterol MDI dispensing events registered for 1984 patients. The number of inhalers a patient obtained at 1 day varied from 1 to 8, but in most cases (83%) only 1 was obtained. However, 9% of all patients received more than 6 albuterol inhalers and 2.9% received more than 12 over the 12 months of observation. As for anti-inflammatory agents, 2363 patients filled a total of 5977 prescriptions during the study period. Most prescriptions were for fluticasone inhalers (20.9%), montelukast tablets (19.9%), theophylline tablets (17.9%), zafirlukast tablets (4.8%), and triamcinolone inhalers (4.5%) (Table 1). 3.2. Quality indicator 1: long-term SABA overuse A total of 6881 SABA dispensing events by 2951 patients were available for analysis of indicator 1. Almost two-thirds of all patients (62.9%) obtained equal or less than a 30-day supply, but 8.7% obtained a supply of more than 180 days in the pharmacies, and 3.4% obtained more than 360 days of supply in the 12-month period. By individual pharmacy, the percentage of long-term SABA overusers varied between 2.2% and 6.1%, equating 7-26 patients per pharmacy (Table 2). Of these, between 1 and 13 patients by pharmacy, were not dispensed any anti-inflammatory asthma medication during the entire period of study. To compare the performance of this quality indicator in smaller time intervals, we calculated quarterly overuse. Quarterly overuse was defined as more than 90 days supply obtained within 1 observed quarter. A total of
437
A.G. Winterstein, A.G. Hartzema / Research in Social and Administrative Pharmacy 1 (2005) 430–445 Table 2 Pharmacy-specific results for indicator 1: long-term SABA overuse
Store no. 1 2 3 4 5 6 7 8
Number of patients with at least 1 SABA dispensing event 209 169 407 534 438 445 428 324
Total
2951
95% CI (based on average n Z 328)
Number of longterm SABA overusers with no antiinflammatory agent
(3.8%) (4.1%) (2.5%) (2.2%) (3.7%) (3.1%) (6.1%) (2.5%)
2.1-6.6 2.3-7.0 1.2-5.0 1.0-4.6 2.0-6.5 1.6-5.8 3.9-9.4 1.2-5.0
2 1 2 3 4 6 13 1
101 (3.4%)
1.7-5.9
Number of long-term SABA overusers (%) 8 7 10 12 16 14 26 8
4424 quarters, in which the 2951 study patients had at least 1 dispensing event, were available for analysis. In 460 (10.4%) of all quarters a patient obtained more than 90 days supply. The number of potential SABA overusers varied between 104 and 126 patients per quarter (Table 3). The denominator is reduced from approximately 3000 patients who obtained a SABA prescription over the 1-year study period to an average of 1000 patients per given quarter. At the same time, the proportion of patients who presumably overused SABA increased when the shorter time interval of 3 months was assessed (between 9.1% and 12.0% per quarter vs 3.4% over the entire year). 3.3. Quality indicator 2: short-term SABA overuse Figure 1 shows the distribution of the time difference (in days) between the number of days supply obtained and the number of days until the patient receives a refill. Negative numbers indicate the number of days a given patient obtained SABA too early, given the daily dose of the medication is limited to the maximum dose of 1 inhaler per month or equivalent. A total of 50.0% of the 3972 refills documented were obtained at Table 3 Number (%) of SABA overusers per study quarter (pooled results of all 8 pharmacies)
Quarter
Number of patients with >1 SABA prescription in the particular quarter
Number (%) of overusers (O90 days supply obtained per given quarter)
1 2 3 4
1080 903 1062 1379
104 108 122 126
(2/00-4/00) (5/00-7/00) (8/00-10/00) (11/00-1/01)
(9.6%) (12.0%) (11.5%) (9.1%)
438
A.G. Winterstein, A.G. Hartzema / Research in Social and Administrative Pharmacy 1 (2005) 430–445 100%
1400
Frequency
1200
80%
75.6%
1000 800
60%
49.8%
600
40%
400
20%
200
0% >200
151-200
101-150
76-100
51-75
26-50
0-25
<0
<-25
<-50
<-75
<-100
0
Days difference (# days supply obtained) - (# days until refill) Figure 1. Distribution of differences (in days) between refill intervals and estimated supply obtained.
least 1 day too early. This is found for 491 (16.6%) of the patients with at least 1 SABA prescription. In 1261 cases refills were obtained in at least 2 consecutive dispensing events too early. At the patient level, for the 2951 patients who filled at least 1 SABA prescription, 217 (7.4%) had at least 1 incident of 2 consecutive early refills (Table 4). More of these short-term overusers received corticosteroids than patients who used SABA in adequate amounts (61.3% vs 29.9%). However, 83 (30.6%) of the 217 identified patients did not receive any corticosteroid during the entire 1-year study period, and 105 patients did not receive any anti-inflammatory agents during the period of SABA overuse. At the pharmacy level, between 3.9% and 11.0% of all patients, who obtained at least 1 SABA prescription in the 1-year study period, had 2 consecutive early refills indicating SABA overuse with the potential for uncontrolled asthma (Figure 2). The 95% confidence intervals illustrate that indicator value differences are expected to be statistically significant (P ! .05) between some pharmacies. 4. Discussion Two indicators were developed to identify asthma patients at increased risk of asthma-related preventable adverse events. Unique to these indicators is their application to individual pharmacies and the inclusion of almost all SABA derivates and dosage forms (in comparison to published indicators based on albuterol or SABA MDIs alone). Including all SABAs and their dosage forms doubled the denominator, and importantly, the crude number of flagged high-risk patients increased from an average of 6 (56 in all 8 pharmacies) to 11 (101 in 8 pharmacies) per pharmacy. Likewise, introduction of the short-term refill-based indicator increases the number of
439
A.G. Winterstein, A.G. Hartzema / Research in Social and Administrative Pharmacy 1 (2005) 430–445
Table 4 Percent (95% CI) of patients with two consecutive too early SABA refills during the 1-year study period 95% confidence Number of patients Number of interval with at least 1 short-term SABA (mean sample Store no. dispensing event overusers (%) size Z 328) 1 2 3 4 5 6 7 8 Total
209 169 407 534 438 445 428 324
15 13 16 35 27 42 51 18
2951
(7.2%) (7.7%) (3.9%) (6.6%) (6.2%) (9.4%) (11.9%) (5.6%)
217 (8.2%)
SABA overusers with no anti-inflammatory dispensing event during period of the 2 early refills
4.7-10.7 5.0-11.1 2.1-6.7 4.3-9.9 3.8-9.3 6.5-13.2 8.6-15.9 3.3-8.5
7 5 6 21 11 22 25 8
4.7-10.7
105
% SABA overusers (95% CI)
potentially uncontrolled patients and offers a mechanism for early detection of exacerbating asthma. In this study, the SABA short-term indicator identified an average of 27 patients per pharmacy who overused SABA, 9 of whom did not receive any anti-inflammatory agent during the 12-month study period. While this triples the numbers identified by the long-term indicator it does not contradict its findings, which require patronage of only 1 pharmacy over long period and chronic (rather than periodically) uncontrolled asthma. Both indicators are expected to have a high–positive predictive value in identifying patients with inappropriate therapy and at high-risk for adverse events for the following 4 reasons. First, the indicators represent the same type of drug exposure that has been shown to be highly predictive of near-fatal and fatal asthma events, regardless of patient diagnosis or other risk factors.10
14 12
11.9
10
9.4
8
7.7
7.2
6
6.6
4
6.2
5.6
3.9
2 0 1
2
3
4
5
6
7
8
Store # Figure 2. Percent (95% CI) of patients with two consecutive ‘‘too early’’ SABA refills during the 1-year study period per pharmacy.
440
A.G. Winterstein, A.G. Hartzema / Research in Social and Administrative Pharmacy 1 (2005) 430–445
Second, the only clinical scenario that would be accompanied by acceptable high-dose SABA use, and thus, not constitute a quality deficit, is chronic obstructive pulmonary disease. However, chronic obstructive pulmonary disease is rare and is preferably treated with anticholinergic agents or SABA combinations thereof, which were excluded in our analysis. Moreover, none of the above-quoted indicators or any of the underlying epidemiologic studies that detected increased mortality rates associated with overuse required a diagnosis of asthma (relative risk estimates range between 1.5 and 5.0 depending on study design and methodology).10,22,23 The overuse of SABA found in these studies was positively associated with dramatically increased morbidity and mortality even if a small proportion of patients who may have used SABA for different indications were included in the analysis. Third, it is unlikely that SABAs were repeatedly obtained but not consumed. Alternative reasons for obtaining such high amounts, such as travel preparations, substitution for lost inhalers, or excessive storage, are unlikely because of the long time frame in indicator 1 or the requirement for repetition in indicator 2. Moreover, patient patronage of different pharmacies becomes more likely when larger periods are observed, which would underestimate and not overestimate the number of patients at risk. Fourth, the average of 27 inappropriately treated patients per pharmacy matches national estimates if broken down at the level of individual pharmacies. In 2000 there were approximately 1500 chain and independent pharmacies registered in North Carolina, which served approximately 8.1 million people.24 For the same year, North Carolina recorded 8018 hospital discharges with a principal diagnosis for asthma (ICD-9-CM 493.xx).25 The National Ambulatory Medical Care Survey reported for the entire U.S. 1.8 Mio. emergency department visits with a primary ICD-9 code for asthma (493.xx) in 2000, which corresponds to 67 emergency department visits per 10 000 people directly related to asthma.26 Approximately 5.3 asthma-induced hospital admissions and 36 emergency department visits per pharmacy would be observed based on these estimates in North Carolina. Thus, our finding of 27 patients per pharmacy with SABA overuse appears valid. Interesting for the quality improvement programs, the short-term indicator can detect variation between pharmacies. For example, the indicator rates for pharmacy numbers 3 and 7 are statistically significantly different (c2 Z 14.3, df Z 1, P ! .001). The long-term indicator proposes this trend as well, but the small differences result in less power for statistical inferences to be made. Likewise, when evaluating the effect of pharmaceutical care in a single pharmacy over time, assuming the sample size of n Z 328 and a prepost design, improvements in indicator rates can be identified as statistically significant, if a 50% reduction (eg, a drop from 8.2% to 4.1%, c2 Z 4.5, df Z 1, P ! .05) was observed. Finally, it is important to note that there is no ‘‘optimal’’ amount of SABA use, ie, patients who use less than 1 inhaler per month most likely do not ‘‘underuse’’ SABA – they have no need. Seven or eight puffs of albuterol
A.G. Winterstein, A.G. Hartzema / Research in Social and Administrative Pharmacy 1 (2005) 430–445
441
per day (which is considered ‘‘normal’’, ie, acceptable dose for our indicator calculations) characterize a patient with moderate to severe asthma. While repeated early refills of 1 or 2 days may appear minor they are clinically important because patients have not exceeded a ‘‘normal’’ but rather the maximum SABA dose. Only 30% of these patients receive anti-inflammatory agents even though appropriate anti-inflammatory therapy was defined very broadly (eg, inclusion of oral corticosteroids). The lack or underuse of inhaled corticosteroids or other anti-inflammatory agents would be another important quality issue that could be addressed by a second set of pharmacy-based asthma quality indicators. Pharmacists can use these indicators in 2 ways. The indicators can be applied to pharmacy information systems on a regular (eg, quarterly) scheduled interval to flag high-risk patients for follow-up. The short-term indicator is most helpful as it allows for early detection of sudden or seasonal asthma deterioration and timely follow-up. Tailored alerts like these asthma indicators could focus the pharmaceutical care services on patients in greatest need and establish a standard in implementing these services. Secondly, indicator rates can be used to monitor one’s own performance in improving quality. Peer pharmacies with better indicator rates and similar patient populations may provide suitable goals for quality improvement. The longitudinal comparison of indicator levels, in turn, provides excellent feedback about the effectiveness of quality improvement initiatives. Both, benchmarking and regular feedback are core principles of quality improvement, which are widely used by hospitals and other provider groups. Pharmacy should follow and demonstrate that the quality of pharmaceutical care is managed in a similar fashion. 4.1. Generalizability and limitations The study sample of 8 pharmacies belonging to the same chain was small, proposing limited generalizability, but there is no reason to believe that the identified drug therapy problems are not applicable to all community pharmacies. In 1998, North Carolina had forecasted self-reported asthma prevalence rates of 5.9% with little variation between counties; for the United States, state-specific prevalence rates ranged between 5.8% and 7.2%.4 Furthermore, SABA overuse has been described as a problem of pharmacotherapy quality throughout the United States and other industrialized countries, thus pharmacists should expect to see similar pattern of inappropriate drug therapy in their pharmacy. The published evidence supports widespread SABA overuse. It further describes variation in SABA overuse across different health care environments, which are interpreted as variation in health care quality. Our indicator findings suggest that the detected variation of indicators rates across pharmacies was not attributable to sampling error, and the observed
442
A.G. Winterstein, A.G. Hartzema / Research in Social and Administrative Pharmacy 1 (2005) 430–445
differences are likely to be real. We assume that the difference in indicator rates of the individual pharmacies reflect variation in health care quality ((a) pharmaceutical care provided, (b) quality of health care provided by other health care professionals, or (c) both), but a formal validation of the indicators against medical charts remains. Even though asthma represents a disease state where the relationship between pharmacotherapy and health outcomes is strong, the predictive validation of indicators against a formal clinical assessment of asthma severity was not done. The evidence of SABA overuse, increased health care utilization, and increased mortality rates was provided by pharmacoepidemiologic studies; the same studies used by national organizations (eg, NHLBI) to develop treatment guidelines.10 These studies used aggregated pharmacy claims data, while this study explores the same at the pharmacy level. Error introduced by the focus on single pharmacies would only apply to the underuse of anti-inflammatory agents, but it is unlikely that patients regularly visit a particular pharmacy to fill their SABA prescriptions and obtain their anti-inflammatory medications in another store. From this we can conclude that our estimate is conservative because patronage of different pharmacies and samples provided by physicians may increase the total SABA amount available to a patient although this is not documented in our database. Finally, we did not consider the use of salmeterol and other long-acting beta-agonists. Even though they are not used for immediate relief, these drugs indicate the presence of asthma symptoms and cannot be used to substitute corticosteroids in patients with persistent asthma.27 The NHLBI guidelines recommend the use of not more than 1 salmeterol inhaler per month, and increase the recommended period a SABA inhaler should last when used concomitantly with salmeterol to 2 months. Both recommendations are not considered in this analysis and might identify additional asthma patients with poor symptom control. Pharmacies have a great opportunity to reduce emergency department visits and hospital admissions. In fact, pharmacies might be able to demonstrate cost savings of $10 000 for 2 avoided hospital admissions (based on an estimated average of $5000 for asthma-related admissions)5 if less than 30 high-risk patients were targeted for pharmaceutical care. As this information is readily available to each pharmacy, a 50% reduction in indicator rates would be an excellent goal if the IOM recommendations for patient safety were being taken seriously.1
5. Conclusion Development and application of pharmacy-specific quality of care indicators is feasible and delivers meaningful data for quality improvement. The developed indicators identify deficits in the quality of care of asthma
443
A.G. Winterstein, A.G. Hartzema / Research in Social and Administrative Pharmacy 1 (2005) 430–445
patients, and allow timely identification of high-risk asthma patients with short-acting beta-agonist overuse. Targeting these patients would offer an excellent opportunity to reduce inappropriate care, make significant contributions to patient and public health, and demonstrate the value of pharmaceutical care.
Appendix 1. SABA and conversion formulas for noninhaler formulations SABA inhalerÒ
Conversion formula
30 days equivalent
2 puffs Z 1 unit-dose vial (3 ml) albuterol nebulizer solution 0.083% (2.5 mg/3 ml) 2 puffs Z 0.5 ml albuterol nebulizer solution 0.5% 2 puffs Z 1 albuterol inhalation capsule (rotacap) 2 puffs Z 2 mg (5 ml) albuterol syrup
1 inhaler Z 100 unit-doses (300 ml)
Metaproterenol inhaler 14 g Z 200 puffs Pirbuterol inhaler 14 g AutohalerÒ Z 400 puffs or 26 g Z 300 puffs
1 inhaler Z 50 ml 1 inhaler Z 100 rotacaps 1 inhaler Z 500 ml 1 inhaler 1 inhaler
Appendix 2. Computation for indicator 3: consecutive short-term overuse of SABA
Dispensing date
Study quarter
Patient 1a 2/10/00 3/6/00 4/4/00 5/1/00 6/1/00 6/26/00 8/8/00 9/11/00 10/9/00 11/9/00 12/8/00 1/8/01
1 1 1 2 2 2 3 3 3 4 4 4
Medication Albuterol Albuterol Albuterol Albuterol Albuterol Albuterol Albuterol Albuterol Albuterol Albuterol Albuterol Albuterol
MDI MDI MDI MDI MDI MDI MDI MDI MDI MDI MDI MDI
Days Amount between refill (A) (g) 17 17 17 17 17 17 17 17 17 17 17 17
* 25 29 27 31 25 43 34 28 31 29 31
Days supply Two obtained A2- consecutive (B) B1 early refills 30 30 30 30 30 30 30 30 30 30 30 30
* ÿ5 ÿ1 ÿ3 1 ÿ5 13 4 ÿ2 1 ÿ1 1
U U
444
A.G. Winterstein, A.G. Hartzema / Research in Social and Administrative Pharmacy 1 (2005) 430–445
Appendix 2. (continued)
Dispensing date
Study quarter
Patient 2 2/1/00
1
3/9/00 3/28/00 4/26/00
1 1 1
Medication
Amount (g)
Albuterol & 17 and Metaproterenol 14 MDI Albuterol MDI 17 Albuterol MDI 17 Albuterol & 17 and Met MDI 14
Days between refill (A)
Days supply obtained (B)
A2B1
*
60
*
37 19 29
30 30 60
ÿ23 ÿ11 ÿ1
Two consecutive early refills
U U
a The first patient received 12 albuterol inhalers during the 12-month study period, which would not be considered long-term overuse. However, the first, second, and third refills occur 5, 1, and 3 days, respectively, too early, which would be considered short-term overuse. The following early refill (ÿ5) is compensated by the next, which occurs 13 days later than expected, if maximum usage is assumed. The second patient receives metaproterenol and albuterol at a single dispensing event. The total amount of supply obtained is summarized in a total of 60 days supply obtained for each of these events.
References 1. Committee on Quality of Health Care in America, Institute of Medicine. To Err is HumandBuilding a Safer Health System. Washington, DC: National Academy Press; 1999. 2. Kanjanarat P, Winterstein AG, Johns TE, Hatton RC, Gonzalez-Rothi R, Segal R. Nature of preventable adverse drug events in hospitals: a literature review. Am J Health Syst Pharm. 2003;60:1750–1759. 3. Winterstein AG, Sauer BC, Hepler CD, Poole C. Preventable drug-related hospital admissions. Ann Pharmacother. 2002;36:1238–1248. 4. Center of Disease Control and Prevention. Forecasted state-specific estimates of selfreported asthma prevalence – United States, 1998. MMWR. 1998;47(47):1022–1025. 5. H.C.U.P. Databases. Healthcare Cost and Utilization Project (HCUP). Agency for Healthcare Research and Quality; 2004. Accessed on October 26, 2004. 6. U.S. Department of Health and Human Services NIoH. National Heart, Lung, and Blood Institute Data Fact Sheet, Asthma Statistics; 1999. 7. Suissa S, Ernst P, Benayoun S, Baltzan M, Cai B. Low-dose inhaled corticosteroids and the prevention of deaths from asthma. N Engl J Med. 2000;343:332–336. 8. Donahue JG, Weiss ST, Livingston JM, Goetsch MA, Greineder DK, Platt R. Inhaled steroids and the risk for hospitalization for asthma. JAMA. 1997;277:887–891. 9. Ernst P, Spitzer WO, Suissa S. Risk of fatal and near-fatal asthma in relation to inhaled corticosteroid use. JAMA. 1992;268:3462–3464. 10. Spitzer WO, Suissa S, Ernst P. The use of beta-agonists and the risk of death and near deaths from asthma. N Engl J Med. 1992;327:1413–1419. 11. National Heart LaBI. Guidelines for the diagnosis and management of asthma: expert panel report No. 2. National Heart LaBI, eds. N.I.H. publication 97-4051. Bethesda, MD; 1997. 12. Enright PL, McClelland RL, Newman AB, Gottlieb DJ, Lebowitz MD. Underdiagnosis and undertreatment of asthma in the elderly. Chest. 1999;116(3):603–613.
A.G. Winterstein, A.G. Hartzema / Research in Social and Administrative Pharmacy 1 (2005) 430–445
445
13. Diette GB, Wu AW, Skinner EA, Markson L, Clark RD, McDonald RC, et al. Treatment patterns among adult patients with asthma: factors associated with overuse of inhaled betaagonists and underuse of inhaled corticosteroids. Arch Intern Med. 1999;159(22):2697– 2704. 14. Guide to prevention quality indicators: hospital admission for ambulatory care sensitive conditions [revision 3]. (AHRQ Pub. no. 02-R0203). Agency for Healthcare Research and Quality (AHRQ). Accessed January 9, 2004. 15. Office of Clinical Standards and Quality QIG. Proposed Summary of Draft 8th Statement of Work. Center of Medicare and Medicaid Services; 2004 Accessed October 26, 2004. 16. The role of the pharmacist in improving asthma care [NIH Publication No. 95-3280]. Bethesda, MD: National Institutes of Health, National Heart, Lung, and Blood Institute; 1995. 17. http://www.qualitytools.ahrq.gov/qualityreport/April29thfinal/ch2qg.doc. Accessed January 18, 2005. 18. Prevention Quality Indicators. Version 2.1. Agency for Healthcare Research and Quality; 2001. 19. CONQUEST Fact Sheet. COmputerized Needs-Oriented QUality Measurement Evaluation SysTem. Rockville, MD: Agency for Health Care Policy and Research; 1999. 20. Malmstron K, Rodriguez-Gomez G, Guerra J, Villaran C, Pineiro A, Wei L. Oral montelukast, inhaled beclomethasone, and placebo for chronic asthma: a randomized, controlled trial. Ann Intern Med. 1999;130(6):487–495. 21. National Committee for Quality Assurance (NCQA). HEDIS 2004. Health plan employer data & information set. In: Technical Specifications, Vol. 2. Washington, DC: National Committee for Quality Assurance (NCQA); 2003. 22. Shireman TI, Heaton PC, Gay WE, Cluxton RJ, Moomaw CJ. Relationship between asthma drug therapy and healthcare utilization. Ann Pharmacother. 2002;36(4):557–564. 23. Piecoro LT, Potoski M, Talbert JC, Doherty DE. Asthma prevalence, cost, and adherence with expert guidelines on the utilization of health care services and costs in a state Medicaid population. Health Serv Res. 2001;362:357–371. 24. North Carolina Board of Pharmacy – Census of registrants. North Carolina Board of Pharmacy; 2002. 25. HCUPnet, Healthcare Cost and Utilization Project. Agency for Heathcare Research and Quality. State Statistics 2000. Rockville, MD: Agency for Healthcare Research and Quality; 2002. 26. McCaig LF, Ly N. National Hospital Ambulatory Medical Care Survey: 2000 Emergency Department Summary. Advance Data, NCHS 326. Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Health Statistics; 2002. 27. Lazarus SC, Boushey HA, Fahy JV, Chinchilli VM, Lemanske RF, Sorkness CA, et al. Long-acting beta-agonist monotherapy vs continued therapy with inhaled corticosteroids in patients with persistent asthma. JAMA. 2001;285:2583–2593.
Pharmacy-specific quality indicators for asthma therapyq Almut G. Winterstein, Ph.D.*, Abraham G. Hartzema, Pharm.D., M.S.P.H., Ph.D.1 College of Pharmacy, Department of Pharmacy Healthcare Administration, University of Florida, PO Box 100496, Gainesville, FL 32610-0496, USA
Abstract Background: Lack of appropriate drug therapy in asthma patients is a recognized quality problem leading to preventable emergency room visits, hospitalizations, or death. While indicators measuring pharmacotherapy quality on the level of prescribers and third party payers are widely used, no such indicators exist for pharmacies. Objective: This study aimed to (1) develop quality indicators for asthma care applicable to retail pharmacies, (2) estimate the prevalence of inappropriate asthma drug therapy, and (3) explore variation in the quality of care across pharmacies. Methods: We present a descriptive analysis of automated patient-specific dispensing data from February 1, 2000 to January 31, 2001 of beta-agonists and antiinflammatory agents with a Food and Drug Administration–approved indication for asthma. Two quality indicators of potentially inappropriate drug therapy were applied: (1) the proportion of patients who obtained more than a 360-day supply
q This work was supported by a grant from the International Federation of Pharmacy (FIP) Foundation. Results were presented in part at the International Federation of Pharmacy Annual Meeting in Vienna, September 2002 and The American Pharmacy Association Annual Meeting in New Orleans, March 2003. * Corresponding author. Tel.: C1 352 273 6258; fax: C1 352 273 6270. E-mail addresses: [email protected]fl.edu (A.G. Winterstein), [email protected]fl.edu (A.G. Hartzema). 1 Perry A. Foote Chair in Health Outcomes Research and Pharmacoeconomics, University of Florida, USA.
1551-7411/$ - see front matter Ó 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.sapharm.2005.07.001
A.G. Winterstein, A.G. Hartzema / Research in Social and Administrative Pharmacy 1 (2005) 430–445
431
(assuming maximum acceptable daily dose) of short-term beta-agonists (SABA) including all inhaler types, inhaler solutions, and syrups during the 12-month study and (2) the proportion of patients with 2 consecutive early refills of defined SABAs. Indicator values with 95% confidence intervals are reported for each pharmacy. Results: Pharmacies had an average of 328 (range 169-534) patients who received SABAs. An average of 11 patients per pharmacy (3.4% of all patients who received SABAs) met the indicator 1 definition. The second indicator identified 8.2% (27) patients per pharmacy as short-term SABA overusers (range 3.9-11.9%). Of these, 48% did not receive any anti-inflammatory agents during the time frame when SABA overuse occurred. Conclusion: Application of drug therapy quality indicators at the level of individual pharmacies using dispensing data is feasible and identifies opportunities for quality improvement. Indicator 2 is most appropriate for daily practice, because it allows for timely identification of potentially uncontrolled patients, and offers a balance between indicator sensitivity and positive predictive value. Ó 2005 Elsevier Inc. All rights reserved. Keywords: Asthma; Quality indicators; Community pharmacy
1. Introduction Research over the past years suggests that suboptimal drug therapy may be one of the most prevalent components of medical care that causes preventable patient injury.1 While most medication error studies have focused on the acute care environment,2 enough is known to conclude that inappropriate drug therapy in ambulatory care contributes significantly to preventable morbidity and mortality.3 Among the high-risk areas for preventable adverse drug events in ambulatory care is the management of asthma. Asthma is responsible for approximately 1.1% of all hospitalizations, 1.5 million emergency department visits, and more than 5500 deaths in the United States annually.4,5 The rate of age-adjusted mortality for asthma has increased steadily despite significant advances in the available pharmacotherapeutic treatment.6 Randomized controlled trials and observational studies demonstrate that the adequate use of inhaled corticosteroids controls asthma symptoms and reduces asthma-related hospital admissions and deaths.7-9 Overuse of betaagonist inhalers, on the other hand, indicates a greater risk for asthmarelated death and near death.10 These findings have directed current treatment guidelines, which emphasize the greater use of anti-inflammatory medications and the judicious use of beta-agonists for short-time symptom relief.11 Despite these well-grounded guidelines, pharmacotherapy in asthma patients is not optimal. Various studies have described quality deficits in asthma care, including lack of diagnosis and treatment, overuse of
432
A.G. Winterstein, A.G. Hartzema / Research in Social and Administrative Pharmacy 1 (2005) 430–445
beta-agonists, and underuse of inhaled corticosteroids.12,13 It is therefore not surprising that the set of patient safety indicators forwarded by the Agency for Healthcare Research and Quality includes the rate of asthmarelated hospital admissions as one of its core measure of health care quality in the United States.14 Quality deficits in asthma treatment as well as other areas in medical care are measured by quality indicators, which are used to assess and benchmark health care quality at the provider, payer, or regional level. Indicator results are summarized in quality scores or report cards and increasingly influence accreditation procedures, consumer choices, and provider reimbursement. Quality assessments by federal constituencies focus traditionally on the hospital and institutional sectors, but the imminent prescription drug benefit for Medicare beneficiaries is expected to add drug utilization evaluations in ambulatory care.15 While pharmacies have the legal mandate for patient counseling and prospective medication reviews, these services are generally excused from formal quality assessments (eg, by the Joint Commission on Accreditation of Healthcare Organizations [JCAHO]). For example, the National Heart, Lung, and Blood Institute (NHLBI) guidelines on ‘‘the role of the pharmacist in improving asthma care’’ define specific tasks a pharmacist should perform. These tasks include patient education, instruction about inhaler techniques and peak flow use, and monitoring medication use.16 However, no specific quality indicator exists that would allow pharmacists to evaluate the quality of care for asthma patients at the level of their individual pharmacy, even though these indicators are available and used at the level of third party payer or physician groups. The development of such quality indicators would allow pharmacists to (1) evaluate and compare the current level of pharmaceutical care provided across pharmacies, and (2) identify and monitor the progress of pharmacybased quality improvement initiatives targeted toward uncontrolled asthma patients.
2. Methods The objectives of this study were to (1) develop quality indicators for asthma care that are applicable to retail pharmacies, (2) estimate the prevalence of inappropriate drug therapy in asthma patients per pharmacy, and (3) explore variation in the quality of care across pharmacies. The developed indicators were expected to meet the following conditions of feasibility and validity.17 Data should be easy to collect and interpret, facilitating initiatives for continuous quality improvement and monitoring change over time. Further, the method selected for data collection should be reliable (replicable and standardized). The quality indicators should be valid in detecting substandard asthma pharmacotherapy, and sensitive to detect
A.G. Winterstein, A.G. Hartzema / Research in Social and Administrative Pharmacy 1 (2005) 430–445
433
variation, ie, to identify differences between pharmacies or to detect changes over time. In addition, the indicators should detect substandard therapy highly predictable of adverse drug events, ie, improvements in indicator rates should likely improve patient outcomes. Two types of quality indicators, outcome-based and process-based, were identified from the literature. Outcome-based indicators focused on health care utilization associated with adverse events (emergency room visits or hospital admissions) that suggest uncontrolled asthma.14,18 Outcome-based indicators were not explored for 3 reasons. First, while these indicators deliver important information about asthma-related outcomes, they represent only indirectly the care processes that were inappropriate and preceded these outcomes; second, patient outcome data on the level of community pharmacies are difficult to retrieve; and third, even though the incidences of ER visits and hospital admissions at the national level are sizeable, they are expected to translate only in very small numbers of affected patients within a single pharmacy. Two indicators reflecting the quality of pharmacotherapy and based on pharmacy dispensing data were defined in this study. Indicator definitions presented follow the Agency of Healthcare Research and Quality criteria. These criteria include documentation of the clinical rationale with evidence sources supporting the measures, the operational description of the denominator and numerator, data sources and data-ascertainment detail, and indicator type.19 Both indicators are based on a cohort study by Spitzer et al10 that demonstrated a 2-fold higher risk for fatal and near-fatal events in patients who use more than 1 albuterol inhaler per month. The current NHLBI guidelines have adopted these findings and expanded their definition of overuse by limiting all short-acting beta-agonist inhalers to 1 per month.11 The study database was created by retrospective extraction of medication records from 8 randomly selected pharmacies of a major pharmacy chain in North Carolina. Extracted database entries included all beta-agonists and anti-inflammatory agents, with a Food and Drug Administration–approved indication for asthma, available in the U.S. market and dispensed between February 1, 2000 and January 31, 2001. Medications were identified by an NDC code and additional full text search using all brand and generic names. Ipratropium bromide and its combinations with albuterol were excluded. Entries related to nasal sprays, creams, ointments, lotions, and ophthalmic solutions were deleted after initial data extraction. A list of the considered medications, beta-agonists and anti-inflammatory agents, is presented in Table 1. For each database entry the following information was available: a unique patient and pharmacy identifier, drug name, dispensed quantity, and dispensing date. The first indicator used a 1-year time window to define long-term overuse of short-term beta-agonists (SABAs). All patients with at least 1 SABA
434
A.G. Winterstein, A.G. Hartzema / Research in Social and Administrative Pharmacy 1 (2005) 430–445
Table 1 Dispensing frequencies of beta-agonists and anti-inflammatory agents in 8 pharmacies between Feb 2000 and Jan 2001 (pooled results) No. of database entries
%
No. of patients
Beta-agonists Terbutaline inhalers Metaproterenol syrupa Terbutaline tabletsa Albuterol inhalation capsules Metaproterenol solution for inhalationa Pirbuterol inhalers Albuterol tablets (including sustained release)a Albuterol syrup Metaproterenol inhalers Salmeterol inhalers and inhalation capsulesa Albuterol solution for inhalation Albuterol inhaler
0 1 10 14 14 247 329 585 693 989 1352 4627
Total
8861
Anti-inflammatory agents Dexamethasone elixir/solution for inhalation Flunisolide inhaler Zileuton tablets Cortisone/hydrocortisone tablets Nedocromil inhaler Dyphylline/Guaifenesin tablets Fludrocortisone tablets Methylprednisolone tablets Budesonide inhaler/solution for inhalation Beclomethason inhaler Cromoglycin inhaler/solution for inhalation Dexamethasone tablets Triamcinolone inhaler Zafirlukast tablets Prednisolone syrup/solution/tablets Theophylline tablets/elixir Montelukast tablets Fluticasone inhaler/rotadisk
8 9 11 21 30 49 101 103 112 199 235 260 271 285 754 1083 1192 1254
0.1 0.2 0.2 0.3 0.5 0.8 1.7 1.7 1.9 3.3 4.0 4.4 4.5 4.8 12.7 18.1 19.9 21.0
8 9 3 5 8 23 22 100 62 88 113 148 120 53 580 198 314 509
Total
5977
100.0
2363
a
0 0.0 0.1 0.2 0.2 2.8 3.7 6.6 8.0 11.1 15.2 52.1 100
0 1 6 4 3 110 103 446 176 301 620 1984 3754
% 0 0.1 0.2 0.1 0.1 2.9 2.7 11.9 4.7 8.0 16.5 52.8 100
0.3 0.4 0.1 0.2 0.3 1.0 0.9 4.2 2.6 3.7 4.8 6.3 5.1 2.2 24.5 8.4 13.3 21.5 100
Not included in the definition of SABA for quality indicator development.
dispensing event in the 1-year period were included in the analysis. Overusers were defined as patients receiving more than 360 days supply within 12 months. All SABA derivates and dosage forms available in the U.S. market except long-acting beta-agonists, sustained release forms (salmeterol and all albuterol and terbutaline tablets), and combination products (eg, CombiventÒ) were included in the analysis. For each dosage form the dispensed amount was converted to the equivalent amount of days
A.G. Winterstein, A.G. Hartzema / Research in Social and Administrative Pharmacy 1 (2005) 430–445
435
of supply assuming the patient required the highest acceptable dose. Each albuterol, metaproterenol, or pirbuterol inhaler was counted as a 30-day supply, which equals the maximum monthly amount of SABA that should be used. Albuterol nebulizer solution, syrup, and rotacaps were transformed into therapeutic equivalent amounts of the respective inhaler formulation. These dosage forms were included because they constituted more than 30% of all filled beta-agonist prescriptions, and were therefore important for detecting overuse. Metaproterenol syrup and nebulizer solution were excluded from analysis because of their small dispensing frequencies. Appendix 1 illustrates drugs and dosage forms included in the analysis and, where applicable, the transformation in inhaler equivalents. Whenever multiple SABA formulations were dispensed at the same day, the dispensed doses were collapsed in a single dispensing event and expressed as the ‘‘total amount of days supply obtained’’. For example, when a patient obtained at a single day 150 unit-doses albuterol nebulizer solution 0.083% (Z45 days supply) and 1 albuterol metered-dose inhaler (30 days supply), a total of 75 days supply was obtained at this particular dispensing encounter. For each patient, the total number of days supply obtained at each dispensing event was summarized for the 1-year study period. The second indicator was based on a shorter time window and expressed as consecutive short-term overuse of SABAs. The denominator included all patients with at least 1 SABA dispensing event. Overuse was defined as 2 consecutive early SABA refills, where early refill is defined as a negative difference between the actual and predicted refill period. Refill intervals were analyzed to facilitate recognition of episodic uncontrolled asthma (occurring over periods shorter than 1 year). The principles for calculating indicator values followed those for calculating medication possession ratios, where the theoretical number of days supply obtained (based on the maximum allowable dose and the total volume dispensed) is divided by the actual number of days until a refill occurs. For this study, we calculated medication possession differences, in which the number of days supply obtained at one particular dispensing event is subtracted from the number of days until a refill occurred. Early refill is defined as any negative difference. Because early refills might occur independent of asthma control (eg, patient happens to stop by the pharmacy or obtains larger supply due to travel plans), short-term SABA overuse was defined as the occurrence of two consecutive early refills. Appendix 2 illustrates the necessary computations and the indicator criteria. We further tested whether the identified SABA overusers received any anti-inflammatory agent during the defined period of overuse. The supporting clinical rationale was derived from several studies, demonstrating an association between anti-inflammatory therapy and reduced morbidity and mortality rates of asthma, as well as the current NHLBI treatment guidelines, recommending the use of anti-inflammatory agents as soon as regular SABA use occurs.7,8,20,21
436
A.G. Winterstein, A.G. Hartzema / Research in Social and Administrative Pharmacy 1 (2005) 430–445
Indicator values were calculated using MS Access 2000Ò and expressed in percent with 95% confidence intervals. Exact confidence intervals (binomial distribution) were calculated using the average sample size of all stores rather than store-specific sample sizes to account for the variation in sample size by store. 3. Results 3.1. Descriptive data analysis The database included 8861 prescriptions of beta-agonists dispensed to a total of 3754 patients (Table 1). More than three-quarters of all filled prescriptions were for albuterol. The number of prescriptions for betaagonists varied across pharmacies. In the 12-month period, between 504 and 1584 beta-agonist prescriptions were filled in the 8 study pharmacies, but the distribution of dispensing frequencies for the different asthma drugs and dosage forms were similar across pharmacies. The largest variation occurred in the proportion of entries for albuterol metered-dosed inhalers (MDI), which varied between 46.0% and 60.7% across pharmacies, and salmeterol inhalers, which varied between 6.4% and 17.4%. The 8 study pharmacies had 4627 albuterol MDI dispensing events registered for 1984 patients. The number of inhalers a patient obtained at 1 day varied from 1 to 8, but in most cases (83%) only 1 was obtained. However, 9% of all patients received more than 6 albuterol inhalers and 2.9% received more than 12 over the 12 months of observation. As for anti-inflammatory agents, 2363 patients filled a total of 5977 prescriptions during the study period. Most prescriptions were for fluticasone inhalers (20.9%), montelukast tablets (19.9%), theophylline tablets (17.9%), zafirlukast tablets (4.8%), and triamcinolone inhalers (4.5%) (Table 1). 3.2. Quality indicator 1: long-term SABA overuse A total of 6881 SABA dispensing events by 2951 patients were available for analysis of indicator 1. Almost two-thirds of all patients (62.9%) obtained equal or less than a 30-day supply, but 8.7% obtained a supply of more than 180 days in the pharmacies, and 3.4% obtained more than 360 days of supply in the 12-month period. By individual pharmacy, the percentage of long-term SABA overusers varied between 2.2% and 6.1%, equating 7-26 patients per pharmacy (Table 2). Of these, between 1 and 13 patients by pharmacy, were not dispensed any anti-inflammatory asthma medication during the entire period of study. To compare the performance of this quality indicator in smaller time intervals, we calculated quarterly overuse. Quarterly overuse was defined as more than 90 days supply obtained within 1 observed quarter. A total of
437
A.G. Winterstein, A.G. Hartzema / Research in Social and Administrative Pharmacy 1 (2005) 430–445 Table 2 Pharmacy-specific results for indicator 1: long-term SABA overuse
Store no. 1 2 3 4 5 6 7 8
Number of patients with at least 1 SABA dispensing event 209 169 407 534 438 445 428 324
Total
2951
95% CI (based on average n Z 328)
Number of longterm SABA overusers with no antiinflammatory agent
(3.8%) (4.1%) (2.5%) (2.2%) (3.7%) (3.1%) (6.1%) (2.5%)
2.1-6.6 2.3-7.0 1.2-5.0 1.0-4.6 2.0-6.5 1.6-5.8 3.9-9.4 1.2-5.0
2 1 2 3 4 6 13 1
101 (3.4%)
1.7-5.9
Number of long-term SABA overusers (%) 8 7 10 12 16 14 26 8
4424 quarters, in which the 2951 study patients had at least 1 dispensing event, were available for analysis. In 460 (10.4%) of all quarters a patient obtained more than 90 days supply. The number of potential SABA overusers varied between 104 and 126 patients per quarter (Table 3). The denominator is reduced from approximately 3000 patients who obtained a SABA prescription over the 1-year study period to an average of 1000 patients per given quarter. At the same time, the proportion of patients who presumably overused SABA increased when the shorter time interval of 3 months was assessed (between 9.1% and 12.0% per quarter vs 3.4% over the entire year). 3.3. Quality indicator 2: short-term SABA overuse Figure 1 shows the distribution of the time difference (in days) between the number of days supply obtained and the number of days until the patient receives a refill. Negative numbers indicate the number of days a given patient obtained SABA too early, given the daily dose of the medication is limited to the maximum dose of 1 inhaler per month or equivalent. A total of 50.0% of the 3972 refills documented were obtained at Table 3 Number (%) of SABA overusers per study quarter (pooled results of all 8 pharmacies)
Quarter
Number of patients with >1 SABA prescription in the particular quarter
Number (%) of overusers (O90 days supply obtained per given quarter)
1 2 3 4
1080 903 1062 1379
104 108 122 126
(2/00-4/00) (5/00-7/00) (8/00-10/00) (11/00-1/01)
(9.6%) (12.0%) (11.5%) (9.1%)
438
A.G. Winterstein, A.G. Hartzema / Research in Social and Administrative Pharmacy 1 (2005) 430–445 100%
1400
Frequency
1200
80%
75.6%
1000 800
60%
49.8%
600
40%
400
20%
200
0% >200
151-200
101-150
76-100
51-75
26-50
0-25
<0
<-25
<-50
<-75
<-100
0
Days difference (# days supply obtained) - (# days until refill) Figure 1. Distribution of differences (in days) between refill intervals and estimated supply obtained.
least 1 day too early. This is found for 491 (16.6%) of the patients with at least 1 SABA prescription. In 1261 cases refills were obtained in at least 2 consecutive dispensing events too early. At the patient level, for the 2951 patients who filled at least 1 SABA prescription, 217 (7.4%) had at least 1 incident of 2 consecutive early refills (Table 4). More of these short-term overusers received corticosteroids than patients who used SABA in adequate amounts (61.3% vs 29.9%). However, 83 (30.6%) of the 217 identified patients did not receive any corticosteroid during the entire 1-year study period, and 105 patients did not receive any anti-inflammatory agents during the period of SABA overuse. At the pharmacy level, between 3.9% and 11.0% of all patients, who obtained at least 1 SABA prescription in the 1-year study period, had 2 consecutive early refills indicating SABA overuse with the potential for uncontrolled asthma (Figure 2). The 95% confidence intervals illustrate that indicator value differences are expected to be statistically significant (P ! .05) between some pharmacies. 4. Discussion Two indicators were developed to identify asthma patients at increased risk of asthma-related preventable adverse events. Unique to these indicators is their application to individual pharmacies and the inclusion of almost all SABA derivates and dosage forms (in comparison to published indicators based on albuterol or SABA MDIs alone). Including all SABAs and their dosage forms doubled the denominator, and importantly, the crude number of flagged high-risk patients increased from an average of 6 (56 in all 8 pharmacies) to 11 (101 in 8 pharmacies) per pharmacy. Likewise, introduction of the short-term refill-based indicator increases the number of
439
A.G. Winterstein, A.G. Hartzema / Research in Social and Administrative Pharmacy 1 (2005) 430–445
Table 4 Percent (95% CI) of patients with two consecutive too early SABA refills during the 1-year study period 95% confidence Number of patients Number of interval with at least 1 short-term SABA (mean sample Store no. dispensing event overusers (%) size Z 328) 1 2 3 4 5 6 7 8 Total
209 169 407 534 438 445 428 324
15 13 16 35 27 42 51 18
2951
(7.2%) (7.7%) (3.9%) (6.6%) (6.2%) (9.4%) (11.9%) (5.6%)
217 (8.2%)
SABA overusers with no anti-inflammatory dispensing event during period of the 2 early refills
4.7-10.7 5.0-11.1 2.1-6.7 4.3-9.9 3.8-9.3 6.5-13.2 8.6-15.9 3.3-8.5
7 5 6 21 11 22 25 8
4.7-10.7
105
% SABA overusers (95% CI)
potentially uncontrolled patients and offers a mechanism for early detection of exacerbating asthma. In this study, the SABA short-term indicator identified an average of 27 patients per pharmacy who overused SABA, 9 of whom did not receive any anti-inflammatory agent during the 12-month study period. While this triples the numbers identified by the long-term indicator it does not contradict its findings, which require patronage of only 1 pharmacy over long period and chronic (rather than periodically) uncontrolled asthma. Both indicators are expected to have a high–positive predictive value in identifying patients with inappropriate therapy and at high-risk for adverse events for the following 4 reasons. First, the indicators represent the same type of drug exposure that has been shown to be highly predictive of near-fatal and fatal asthma events, regardless of patient diagnosis or other risk factors.10
14 12
11.9
10
9.4
8
7.7
7.2
6
6.6
4
6.2
5.6
3.9
2 0 1
2
3
4
5
6
7
8
Store # Figure 2. Percent (95% CI) of patients with two consecutive ‘‘too early’’ SABA refills during the 1-year study period per pharmacy.
440
A.G. Winterstein, A.G. Hartzema / Research in Social and Administrative Pharmacy 1 (2005) 430–445
Second, the only clinical scenario that would be accompanied by acceptable high-dose SABA use, and thus, not constitute a quality deficit, is chronic obstructive pulmonary disease. However, chronic obstructive pulmonary disease is rare and is preferably treated with anticholinergic agents or SABA combinations thereof, which were excluded in our analysis. Moreover, none of the above-quoted indicators or any of the underlying epidemiologic studies that detected increased mortality rates associated with overuse required a diagnosis of asthma (relative risk estimates range between 1.5 and 5.0 depending on study design and methodology).10,22,23 The overuse of SABA found in these studies was positively associated with dramatically increased morbidity and mortality even if a small proportion of patients who may have used SABA for different indications were included in the analysis. Third, it is unlikely that SABAs were repeatedly obtained but not consumed. Alternative reasons for obtaining such high amounts, such as travel preparations, substitution for lost inhalers, or excessive storage, are unlikely because of the long time frame in indicator 1 or the requirement for repetition in indicator 2. Moreover, patient patronage of different pharmacies becomes more likely when larger periods are observed, which would underestimate and not overestimate the number of patients at risk. Fourth, the average of 27 inappropriately treated patients per pharmacy matches national estimates if broken down at the level of individual pharmacies. In 2000 there were approximately 1500 chain and independent pharmacies registered in North Carolina, which served approximately 8.1 million people.24 For the same year, North Carolina recorded 8018 hospital discharges with a principal diagnosis for asthma (ICD-9-CM 493.xx).25 The National Ambulatory Medical Care Survey reported for the entire U.S. 1.8 Mio. emergency department visits with a primary ICD-9 code for asthma (493.xx) in 2000, which corresponds to 67 emergency department visits per 10 000 people directly related to asthma.26 Approximately 5.3 asthma-induced hospital admissions and 36 emergency department visits per pharmacy would be observed based on these estimates in North Carolina. Thus, our finding of 27 patients per pharmacy with SABA overuse appears valid. Interesting for the quality improvement programs, the short-term indicator can detect variation between pharmacies. For example, the indicator rates for pharmacy numbers 3 and 7 are statistically significantly different (c2 Z 14.3, df Z 1, P ! .001). The long-term indicator proposes this trend as well, but the small differences result in less power for statistical inferences to be made. Likewise, when evaluating the effect of pharmaceutical care in a single pharmacy over time, assuming the sample size of n Z 328 and a prepost design, improvements in indicator rates can be identified as statistically significant, if a 50% reduction (eg, a drop from 8.2% to 4.1%, c2 Z 4.5, df Z 1, P ! .05) was observed. Finally, it is important to note that there is no ‘‘optimal’’ amount of SABA use, ie, patients who use less than 1 inhaler per month most likely do not ‘‘underuse’’ SABA – they have no need. Seven or eight puffs of albuterol
A.G. Winterstein, A.G. Hartzema / Research in Social and Administrative Pharmacy 1 (2005) 430–445
441
per day (which is considered ‘‘normal’’, ie, acceptable dose for our indicator calculations) characterize a patient with moderate to severe asthma. While repeated early refills of 1 or 2 days may appear minor they are clinically important because patients have not exceeded a ‘‘normal’’ but rather the maximum SABA dose. Only 30% of these patients receive anti-inflammatory agents even though appropriate anti-inflammatory therapy was defined very broadly (eg, inclusion of oral corticosteroids). The lack or underuse of inhaled corticosteroids or other anti-inflammatory agents would be another important quality issue that could be addressed by a second set of pharmacy-based asthma quality indicators. Pharmacists can use these indicators in 2 ways. The indicators can be applied to pharmacy information systems on a regular (eg, quarterly) scheduled interval to flag high-risk patients for follow-up. The short-term indicator is most helpful as it allows for early detection of sudden or seasonal asthma deterioration and timely follow-up. Tailored alerts like these asthma indicators could focus the pharmaceutical care services on patients in greatest need and establish a standard in implementing these services. Secondly, indicator rates can be used to monitor one’s own performance in improving quality. Peer pharmacies with better indicator rates and similar patient populations may provide suitable goals for quality improvement. The longitudinal comparison of indicator levels, in turn, provides excellent feedback about the effectiveness of quality improvement initiatives. Both, benchmarking and regular feedback are core principles of quality improvement, which are widely used by hospitals and other provider groups. Pharmacy should follow and demonstrate that the quality of pharmaceutical care is managed in a similar fashion. 4.1. Generalizability and limitations The study sample of 8 pharmacies belonging to the same chain was small, proposing limited generalizability, but there is no reason to believe that the identified drug therapy problems are not applicable to all community pharmacies. In 1998, North Carolina had forecasted self-reported asthma prevalence rates of 5.9% with little variation between counties; for the United States, state-specific prevalence rates ranged between 5.8% and 7.2%.4 Furthermore, SABA overuse has been described as a problem of pharmacotherapy quality throughout the United States and other industrialized countries, thus pharmacists should expect to see similar pattern of inappropriate drug therapy in their pharmacy. The published evidence supports widespread SABA overuse. It further describes variation in SABA overuse across different health care environments, which are interpreted as variation in health care quality. Our indicator findings suggest that the detected variation of indicators rates across pharmacies was not attributable to sampling error, and the observed
442
A.G. Winterstein, A.G. Hartzema / Research in Social and Administrative Pharmacy 1 (2005) 430–445
differences are likely to be real. We assume that the difference in indicator rates of the individual pharmacies reflect variation in health care quality ((a) pharmaceutical care provided, (b) quality of health care provided by other health care professionals, or (c) both), but a formal validation of the indicators against medical charts remains. Even though asthma represents a disease state where the relationship between pharmacotherapy and health outcomes is strong, the predictive validation of indicators against a formal clinical assessment of asthma severity was not done. The evidence of SABA overuse, increased health care utilization, and increased mortality rates was provided by pharmacoepidemiologic studies; the same studies used by national organizations (eg, NHLBI) to develop treatment guidelines.10 These studies used aggregated pharmacy claims data, while this study explores the same at the pharmacy level. Error introduced by the focus on single pharmacies would only apply to the underuse of anti-inflammatory agents, but it is unlikely that patients regularly visit a particular pharmacy to fill their SABA prescriptions and obtain their anti-inflammatory medications in another store. From this we can conclude that our estimate is conservative because patronage of different pharmacies and samples provided by physicians may increase the total SABA amount available to a patient although this is not documented in our database. Finally, we did not consider the use of salmeterol and other long-acting beta-agonists. Even though they are not used for immediate relief, these drugs indicate the presence of asthma symptoms and cannot be used to substitute corticosteroids in patients with persistent asthma.27 The NHLBI guidelines recommend the use of not more than 1 salmeterol inhaler per month, and increase the recommended period a SABA inhaler should last when used concomitantly with salmeterol to 2 months. Both recommendations are not considered in this analysis and might identify additional asthma patients with poor symptom control. Pharmacies have a great opportunity to reduce emergency department visits and hospital admissions. In fact, pharmacies might be able to demonstrate cost savings of $10 000 for 2 avoided hospital admissions (based on an estimated average of $5000 for asthma-related admissions)5 if less than 30 high-risk patients were targeted for pharmaceutical care. As this information is readily available to each pharmacy, a 50% reduction in indicator rates would be an excellent goal if the IOM recommendations for patient safety were being taken seriously.1
5. Conclusion Development and application of pharmacy-specific quality of care indicators is feasible and delivers meaningful data for quality improvement. The developed indicators identify deficits in the quality of care of asthma
443
A.G. Winterstein, A.G. Hartzema / Research in Social and Administrative Pharmacy 1 (2005) 430–445
patients, and allow timely identification of high-risk asthma patients with short-acting beta-agonist overuse. Targeting these patients would offer an excellent opportunity to reduce inappropriate care, make significant contributions to patient and public health, and demonstrate the value of pharmaceutical care.
Appendix 1. SABA and conversion formulas for noninhaler formulations SABA inhalerÒ
Conversion formula
30 days equivalent
2 puffs Z 1 unit-dose vial (3 ml) albuterol nebulizer solution 0.083% (2.5 mg/3 ml) 2 puffs Z 0.5 ml albuterol nebulizer solution 0.5% 2 puffs Z 1 albuterol inhalation capsule (rotacap) 2 puffs Z 2 mg (5 ml) albuterol syrup
1 inhaler Z 100 unit-doses (300 ml)
Metaproterenol inhaler 14 g Z 200 puffs Pirbuterol inhaler 14 g AutohalerÒ Z 400 puffs or 26 g Z 300 puffs
1 inhaler Z 50 ml 1 inhaler Z 100 rotacaps 1 inhaler Z 500 ml 1 inhaler 1 inhaler
Appendix 2. Computation for indicator 3: consecutive short-term overuse of SABA
Dispensing date
Study quarter
Patient 1a 2/10/00 3/6/00 4/4/00 5/1/00 6/1/00 6/26/00 8/8/00 9/11/00 10/9/00 11/9/00 12/8/00 1/8/01
1 1 1 2 2 2 3 3 3 4 4 4
Medication Albuterol Albuterol Albuterol Albuterol Albuterol Albuterol Albuterol Albuterol Albuterol Albuterol Albuterol Albuterol
MDI MDI MDI MDI MDI MDI MDI MDI MDI MDI MDI MDI
Days Amount between refill (A) (g) 17 17 17 17 17 17 17 17 17 17 17 17
* 25 29 27 31 25 43 34 28 31 29 31
Days supply Two obtained A2- consecutive (B) B1 early refills 30 30 30 30 30 30 30 30 30 30 30 30
* ÿ5 ÿ1 ÿ3 1 ÿ5 13 4 ÿ2 1 ÿ1 1
U U
444
A.G. Winterstein, A.G. Hartzema / Research in Social and Administrative Pharmacy 1 (2005) 430–445
Appendix 2. (continued)
Dispensing date
Study quarter
Patient 2 2/1/00
1
3/9/00 3/28/00 4/26/00
1 1 1
Medication
Amount (g)
Albuterol & 17 and Metaproterenol 14 MDI Albuterol MDI 17 Albuterol MDI 17 Albuterol & 17 and Met MDI 14
Days between refill (A)
Days supply obtained (B)
A2B1
*
60
*
37 19 29
30 30 60
ÿ23 ÿ11 ÿ1
Two consecutive early refills
U U
a The first patient received 12 albuterol inhalers during the 12-month study period, which would not be considered long-term overuse. However, the first, second, and third refills occur 5, 1, and 3 days, respectively, too early, which would be considered short-term overuse. The following early refill (ÿ5) is compensated by the next, which occurs 13 days later than expected, if maximum usage is assumed. The second patient receives metaproterenol and albuterol at a single dispensing event. The total amount of supply obtained is summarized in a total of 60 days supply obtained for each of these events.
References 1. Committee on Quality of Health Care in America, Institute of Medicine. To Err is HumandBuilding a Safer Health System. Washington, DC: National Academy Press; 1999. 2. Kanjanarat P, Winterstein AG, Johns TE, Hatton RC, Gonzalez-Rothi R, Segal R. Nature of preventable adverse drug events in hospitals: a literature review. Am J Health Syst Pharm. 2003;60:1750–1759. 3. Winterstein AG, Sauer BC, Hepler CD, Poole C. Preventable drug-related hospital admissions. Ann Pharmacother. 2002;36:1238–1248. 4. Center of Disease Control and Prevention. Forecasted state-specific estimates of selfreported asthma prevalence – United States, 1998. MMWR. 1998;47(47):1022–1025. 5. H.C.U.P. Databases. Healthcare Cost and Utilization Project (HCUP). Agency for Healthcare Research and Quality; 2004. Accessed on October 26, 2004. 6. U.S. Department of Health and Human Services NIoH. National Heart, Lung, and Blood Institute Data Fact Sheet, Asthma Statistics; 1999. 7. Suissa S, Ernst P, Benayoun S, Baltzan M, Cai B. Low-dose inhaled corticosteroids and the prevention of deaths from asthma. N Engl J Med. 2000;343:332–336. 8. Donahue JG, Weiss ST, Livingston JM, Goetsch MA, Greineder DK, Platt R. Inhaled steroids and the risk for hospitalization for asthma. JAMA. 1997;277:887–891. 9. Ernst P, Spitzer WO, Suissa S. Risk of fatal and near-fatal asthma in relation to inhaled corticosteroid use. JAMA. 1992;268:3462–3464. 10. Spitzer WO, Suissa S, Ernst P. The use of beta-agonists and the risk of death and near deaths from asthma. N Engl J Med. 1992;327:1413–1419. 11. National Heart LaBI. Guidelines for the diagnosis and management of asthma: expert panel report No. 2. National Heart LaBI, eds. N.I.H. publication 97-4051. Bethesda, MD; 1997. 12. Enright PL, McClelland RL, Newman AB, Gottlieb DJ, Lebowitz MD. Underdiagnosis and undertreatment of asthma in the elderly. Chest. 1999;116(3):603–613.
A.G. Winterstein, A.G. Hartzema / Research in Social and Administrative Pharmacy 1 (2005) 430–445
445
13. Diette GB, Wu AW, Skinner EA, Markson L, Clark RD, McDonald RC, et al. Treatment patterns among adult patients with asthma: factors associated with overuse of inhaled betaagonists and underuse of inhaled corticosteroids. Arch Intern Med. 1999;159(22):2697– 2704. 14. Guide to prevention quality indicators: hospital admission for ambulatory care sensitive conditions [revision 3]. (AHRQ Pub. no. 02-R0203). Agency for Healthcare Research and Quality (AHRQ). Accessed January 9, 2004. 15. Office of Clinical Standards and Quality QIG. Proposed Summary of Draft 8th Statement of Work. Center of Medicare and Medicaid Services; 2004 Accessed October 26, 2004. 16. The role of the pharmacist in improving asthma care [NIH Publication No. 95-3280]. Bethesda, MD: National Institutes of Health, National Heart, Lung, and Blood Institute; 1995. 17. http://www.qualitytools.ahrq.gov/qualityreport/April29thfinal/ch2qg.doc. Accessed January 18, 2005. 18. Prevention Quality Indicators. Version 2.1. Agency for Healthcare Research and Quality; 2001. 19. CONQUEST Fact Sheet. COmputerized Needs-Oriented QUality Measurement Evaluation SysTem. Rockville, MD: Agency for Health Care Policy and Research; 1999. 20. Malmstron K, Rodriguez-Gomez G, Guerra J, Villaran C, Pineiro A, Wei L. Oral montelukast, inhaled beclomethasone, and placebo for chronic asthma: a randomized, controlled trial. Ann Intern Med. 1999;130(6):487–495. 21. National Committee for Quality Assurance (NCQA). HEDIS 2004. Health plan employer data & information set. In: Technical Specifications, Vol. 2. Washington, DC: National Committee for Quality Assurance (NCQA); 2003. 22. Shireman TI, Heaton PC, Gay WE, Cluxton RJ, Moomaw CJ. Relationship between asthma drug therapy and healthcare utilization. Ann Pharmacother. 2002;36(4):557–564. 23. Piecoro LT, Potoski M, Talbert JC, Doherty DE. Asthma prevalence, cost, and adherence with expert guidelines on the utilization of health care services and costs in a state Medicaid population. Health Serv Res. 2001;362:357–371. 24. North Carolina Board of Pharmacy – Census of registrants. North Carolina Board of Pharmacy; 2002. 25. HCUPnet, Healthcare Cost and Utilization Project. Agency for Heathcare Research and Quality. State Statistics 2000. Rockville, MD: Agency for Healthcare Research and Quality; 2002. 26. McCaig LF, Ly N. National Hospital Ambulatory Medical Care Survey: 2000 Emergency Department Summary. Advance Data, NCHS 326. Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Health Statistics; 2002. 27. Lazarus SC, Boushey HA, Fahy JV, Chinchilli VM, Lemanske RF, Sorkness CA, et al. Long-acting beta-agonist monotherapy vs continued therapy with inhaled corticosteroids in patients with persistent asthma. JAMA. 2001;285:2583–2593.