US Physician Adherence to Standards in Asthma Pharmacotherapy Varies by Patient and Physician Characteristics

Abstract

Asthma, a serious public health problem, is characterized by reversible airway obstruction and hypersensitivity resulting from inflammation of the airways. The prevalence and societal burden of asthma is compelling, as well as disturbing, especially given the availability of effective evidence-based treatment.1Centers for Disease Control and Prevention Surveillance for asthma United States, 1980-1999.MMWR Morb Mortal Wkly Rep. 2002; 51: 1-13Google Scholar According to the National Asthma Education and Prevention Project (NAEPP), adequate asthma control relies on the use of appropriate pharmacotherapy accompanied by several other effective management strategies.2National Asthma Education Program (NAEP) Expert panel report: guidelines for the diagnosis and management of asthma update on selected topics—2002.J Allergy Clin Immunol. 2002; 110: 141-219Abstract Full Text PDF Google Scholar, 3National Asthma Education Program (NAEP) Expert panel report II: guidelines for the diagnosis and management of asthma. National Heart, Lung, and Blood Institute, Bethesda (MD)1997Google Scholar Throughout the 1990s, the emphasis of asthma pharmacotherapy has shifted from relieving acute bronchospasm with short-acting reliever medications (eg, short-acting β2-agonists) to long-term proactive control of the underlying inflammation with anti-inflammatory medications (eg, inhaled steroids).Collective evidence, however, consistently indicates a gap between established guidelines and clinical practice regarding asthma pharmacotherapy, although this gap might be shrinking.4Stafford RS Ma J Finkelstein SN Haver K Cockburn I National trends in asthma visits and asthma pharmacotherapy, 1978-2002.J Allergy Clin Immunol. 2003; 111: 729-735Abstract Full Text Full Text PDF PubMed Scopus (69) Google Scholar Inappropriate pharmacotherapy for asthma leads to increased morbidity and mortality, as well as increased use of health care resources.5Anis AH Lynd LD Wang XH King G Spinelli JJ Fitzgerald M et al.Double trouble: impact of inappropriate use of asthma medication on the use of health care resources.CMAJ. 2001; 164: 625-631PubMed Google Scholar Understanding the factors associated with inappropriate asthma medication use will guide interventions to improve quality of care and patient outcomes. A majority of past studies investigating inappropriate asthma medication use rely on patient self-reports, claims data, and pharmacy records. These data sources cannot directly disentangle physician guideline adherence from prescription dispensing, patient compliance, or both. This study examines patient and physician characteristics associated with physician-reported prescribing of inhaled steroids and short-acting β2-agonists. We hypothesized that visits by non-Hispanic whites and those made to asthma specialists would be associated with greater use of inhaled steroids and less use of short-acting β2-agonists.Study data were obtained from the National Ambulatory Medical Care Survey (NAMCS) and the outpatient department component of the National Hospital Ambulatory Medical Care Survey (NHAMCS). NAMCS captures health care services provided by office-based physicians, whereas NHAMCS surveys practices in hospital outpatient departments. Both surveys, conducted by the National Center for Health Statistics in Hyattsville, Md,6National Center for Health Statistics. Ambulatory Health Care Data. Available at: http://www.cdc.gov/nchs/about/major/ahcd/ahcd1.htm. Accessed June 5, 2003.Google Scholar use multistage probability sampling procedures and enable essentially unbiased national estimates to be made. We combined NAMCS and NHAMCS data from 1995 through 2000 to include a wider range of outpatient settings and a broader socioeconomic spectrum of patients seeking ambulatory care.The unit of analysis in both surveys is patient visits. Statistical analyses were performed by using SUDAAN software (RTI, Research Triangle Park, NC), and the significance level was set at a P value of less than .01. Asthma visits were identified by an International Classification of Disease, Ninth Revision, Clinical Modification code of 493 or an N(H)AMCS-specific reason for visit code of 2625.0, excluding visits of patients with reported chronic obstructive pulmonary disease. Multivariate logistic regression models were developed by using the SUDAAN RLOGIST procedure to assess the relationship of the prescribing of inhaled steroids and short-acting β2-agonists to patient and physician characteristics. Patient characteristics included race-ethnicity, insurance status, age, sex, geographic region, metropolitan area status, and patient visit status. The patient's medical insurance status was classified as private (Blue Cross/Blue Shield or other private or commercial insurance) and nonprivate (Medicare, Medicaid, workers' compensation, other insurance, and self-pay). Patient age was dichotomized to younger than 18 years (children) and 18 years and older (adults). A new patient visit was defined as a patient being seen by the physician for the first time or if the purpose of the visit was for a general medical examination. For the purposes of this study, physician specialty was categorized as generalists (ie, general-family practitioners, internists, and pediatricians), asthma specialists (ie, allergists and pulmonologists), and all others. Physician practice settings included private offices (NAMCS) and hospital outpatient departments (NHAMCS). Among the selected characteristics, patient race-ethnicity and physician specialty are of particular interest. Changes in racial-ethnic and specialty differences over time were examined by defining interaction terms in logistic regression models; however, no significant difference was found.Overall, inhaled steroids were reportedly prescribed during 24% (99% CI, 21% to 26%) and short-acting β2-agonists during 55% (95% CI, 52% to 58%) of all asthma visits. Several characteristics significantly altered the odds of physician-reported prescribing of inhaled steroids and short-acting β2-agonists (Table I). After controlling for other characteristics, the likelihood of visits by Hispanics being prescribed inhaled steroids was 43% less than that in visits by whites, whereas patients cared for by asthma specialists were 3 times more likely to be prescribed inhaled steroids than those seen by generalists. Physician-reported prescribing of short-acting β2-agonists did not differ by patient race-ethnicity or physician specialty. Compared with their adult counterparts, children and adolescent visitors were 43% less likely to be prescribed inhaled steroids but 80% more likely to be prescribed short-acting β2-agonists. Relative to visits to hospital outpatient departments, visitors to in-office physicians were less likely to be prescribed both inhaled steroids (66% lower) and short-acting β2-agonists (61% lower). In addition, asthma visits by patients with private insurance versus those without private insurance had 1.51 times the odds of being prescribed short-acting β2-agonists. Physician-reported prescribing did not vary significantly on the basis of the remaining characteristics.Table IIndependent effects of patient and physician characteristics on physician-reported prescribing of inhaled steroids and short-acting β2-agonists, NAMCS/NHMCS 1995-2000Inhaled steroids, n (%)Adjusted odds ratio∗The odds ratios for each variable were adjusted for all other selected patient and physician characteristics. (99% CI)Short-acting β2-agonists, n (%)Adjusted odds ratio (99% CI)R ace-ethnicityWhite641 (24)1.00 (reference)1503 (56)1.00 (reference)African American295 (26)1.37 (0.80-2.36)765 (57)1.04 (0.68-1.58)Hispanic165 (16)0.57 (0.32-0.99)528 (60)1.03 (0.65-1.64)Other32 (20)0.63 (0.20-1.99)94 (48)0.67 (0.29-1.57)P hysician specialtyGeneralist268 (18)1.00 (reference)846 (57)1.00 (reference)Specialist194 (46)3.69 (2.31-5.89)238 (55)1.07 (0.59-1.95)Other832 (19)0.53 (0.25-1.10)2207 (49)0.43 (0.21-0.87)M edical insurancePrivate497 (27)1.25 (0.86-1.80)1172 (60)1.51 (1.08-2.12)Nonprivate775 (20)1.00 (reference)2074 (50)1.00 (reference)A ge groupChildren482 (18)0.57 (0.37-0.88)1621 (65)1.80 (1.26-2.57)Adults812 (28)1.00 (reference)1670 (49)1.00 (reference)S urvey typeNAMCS499 (24)0.34 (0.16-0.75)1169 (55)0.39 (0.19-0.80)NHAMCS795 (24)1.00 (reference)2122 (59)1.00 (reference)∗ The odds ratios for each variable were adjusted for all other selected patient and physician characteristics. Open table in a new tab The success of asthma treatment hinges on appropriate medication use according to evidence-based practice guidelines, such as those published by NAEPP.2National Asthma Education Program (NAEP) Expert panel report: guidelines for the diagnosis and management of asthma update on selected topics—2002.J Allergy Clin Immunol. 2002; 110: 141-219Abstract Full Text PDF Google Scholar, 3National Asthma Education Program (NAEP) Expert panel report II: guidelines for the diagnosis and management of asthma. National Heart, Lung, and Blood Institute, Bethesda (MD)1997Google Scholar It is evident from this and many past studies that appropriate use of drugs in these 2 dominant treatment classes remains elusive. Many studies have attempted to identify underlying reasons for the continuing discrepancies between current asthma care and evidence-based guidelines. Our study reveals that physician nonadherence to NAEPP guidelines in the use of inhaled steroids and short-acting β2-agonists is not uncommon during ambulatory asthma visits in the United States. Moreover, asthma pharmacotherapy in children, minorities, patients covered by nonprivate insurance, and generalists' patients conforms less well to consensus guidelines. These disparities in asthma care do not appear to have decreased in recent years.Our data preclude the determination of reasons for the observed prescribing variations, but some speculations can be made on the basis of published studies.2National Asthma Education Program (NAEP) Expert panel report: guidelines for the diagnosis and management of asthma update on selected topics—2002.J Allergy Clin Immunol. 2002; 110: 141-219Abstract Full Text PDF Google Scholar, 7Chambers CV Markson L Diamond JJ Lasch L Berger M Health beliefs and compliance with inhaled corticosteroids by asthmatic patients in primary care practices.Respir Med. 1999; 93: 88-94Abstract Full Text PDF PubMed Scopus (164) Google Scholar, 8Vollmer WM O'Hollaren M Ettinger KM Stibolt T Wilkins J Buist AS et al.Specialty differences in the management of asthma. A cross-sectional assessment of allergists' patients and generalists' patients in a large HMO.Arch Intern Med. 1997; 157: 1201-1208Crossref PubMed Google Scholar Apparently less optimal asthma treatment in children might be attributed to concerns about potential adverse effects of inhaled steroids, family-maternal factors, and child-parental health beliefs regarding asthma and the value of prevention. For racial-ethnic minorities, there might be multiple barriers, including access to health care, patient-clinician communication, and cultural differences in health beliefs. Variations in practice among physicians might be due to differences in training and knowledge of the disease and relevant guidelines, as well as in organizational structure and resource allocation. Regardless of the underlying reasons, the observed disparities manifest a practical departure from guidelines, which recommend no differential therapy for any subpopulation other than different dose, number, and frequency of medications by asthma severity.2National Asthma Education Program (NAEP) Expert panel report: guidelines for the diagnosis and management of asthma update on selected topics—2002.J Allergy Clin Immunol. 2002; 110: 141-219Abstract Full Text PDF Google Scholar, 3National Asthma Education Program (NAEP) Expert panel report II: guidelines for the diagnosis and management of asthma. National Heart, Lung, and Blood Institute, Bethesda (MD)1997Google Scholar Addressing these disparities is particularly important in that asthma morbidity can be prevented and use of health care resources can be reduced by appropriate drug treatment.The current findings should be interpreted in the context of several limitations of NAMCS and NHAMCS data. Both data sources are serial cross-sectional surveys of patient visits recorded by health care providers. An inherent limitation of any survey study is response bias; also, reported medication use only reflects a physician's best knowledge of any new and continued prescriptions. In addition, lack of patient-level clinical data limits our assessment of the appropriateness of medication regimens because of the inability to adjust for some known correlates of asthma medication use: drug dosage, asthma severity, symptom frequency, duration and frequency of health care use, and patient compliance in filling and using prescribed medications.Nonetheless, we have no reason to believe that the above data limitations differentially affect subgroups of the study sample according to the selected characteristics so as to compromise the validity of cross-group comparisons. By using data from physician-based national surveys, this study suggests that the prescribing of 2 dominant classes of asthma medications (ie, inhaled steroids and short-acting β2-agonists) varies by nonclinical factors. These factors pinpoint subpopulations of patients with asthma whose drug regimens need to be reassessed and groups of physicians whose practices for treating asthma need to be improved. These findings should be used to guide future interventions that aim to bring physician practice into agreement with published guidelines for optimal asthma treatment. Asthma, a serious public health problem, is characterized by reversible airway obstruction and hypersensitivity resulting from inflammation of the airways. The prevalence and societal burden of asthma is compelling, as well as disturbing, especially given the availability of effective evidence-based treatment.1Centers for Disease Control and Prevention Surveillance for asthma United States, 1980-1999.MMWR Morb Mortal Wkly Rep. 2002; 51: 1-13Google Scholar According to the National Asthma Education and Prevention Project (NAEPP), adequate asthma control relies on the use of appropriate pharmacotherapy accompanied by several other effective management strategies.2National Asthma Education Program (NAEP) Expert panel report: guidelines for the diagnosis and management of asthma update on selected topics—2002.J Allergy Clin Immunol. 2002; 110: 141-219Abstract Full Text PDF Google Scholar, 3National Asthma Education Program (NAEP) Expert panel report II: guidelines for the diagnosis and management of asthma. National Heart, Lung, and Blood Institute, Bethesda (MD)1997Google Scholar Throughout the 1990s, the emphasis of asthma pharmacotherapy has shifted from relieving acute bronchospasm with short-acting reliever medications (eg, short-acting β2-agonists) to long-term proactive control of the underlying inflammation with anti-inflammatory medications (eg, inhaled steroids). Collective evidence, however, consistently indicates a gap between established guidelines and clinical practice regarding asthma pharmacotherapy, although this gap might be shrinking.4Stafford RS Ma J Finkelstein SN Haver K Cockburn I National trends in asthma visits and asthma pharmacotherapy, 1978-2002.J Allergy Clin Immunol. 2003; 111: 729-735Abstract Full Text Full Text PDF PubMed Scopus (69) Google Scholar Inappropriate pharmacotherapy for asthma leads to increased morbidity and mortality, as well as increased use of health care resources.5Anis AH Lynd LD Wang XH King G Spinelli JJ Fitzgerald M et al.Double trouble: impact of inappropriate use of asthma medication on the use of health care resources.CMAJ. 2001; 164: 625-631PubMed Google Scholar Understanding the factors associated with inappropriate asthma medication use will guide interventions to improve quality of care and patient outcomes. A majority of past studies investigating inappropriate asthma medication use rely on patient self-reports, claims data, and pharmacy records. These data sources cannot directly disentangle physician guideline adherence from prescription dispensing, patient compliance, or both. This study examines patient and physician characteristics associated with physician-reported prescribing of inhaled steroids and short-acting β2-agonists. We hypothesized that visits by non-Hispanic whites and those made to asthma specialists would be associated with greater use of inhaled steroids and less use of short-acting β2-agonists. Study data were obtained from the National Ambulatory Medical Care Survey (NAMCS) and the outpatient department component of the National Hospital Ambulatory Medical Care Survey (NHAMCS). NAMCS captures health care services provided by office-based physicians, whereas NHAMCS surveys practices in hospital outpatient departments. Both surveys, conducted by the National Center for Health Statistics in Hyattsville, Md,6National Center for Health Statistics. Ambulatory Health Care Data. Available at: http://www.cdc.gov/nchs/about/major/ahcd/ahcd1.htm. Accessed June 5, 2003.Google Scholar use multistage probability sampling procedures and enable essentially unbiased national estimates to be made. We combined NAMCS and NHAMCS data from 1995 through 2000 to include a wider range of outpatient settings and a broader socioeconomic spectrum of patients seeking ambulatory care. The unit of analysis in both surveys is patient visits. Statistical analyses were performed by using SUDAAN software (RTI, Research Triangle Park, NC), and the significance level was set at a P value of less than .01. Asthma visits were identified by an International Classification of Disease, Ninth Revision, Clinical Modification code of 493 or an N(H)AMCS-specific reason for visit code of 2625.0, excluding visits of patients with reported chronic obstructive pulmonary disease. Multivariate logistic regression models were developed by using the SUDAAN RLOGIST procedure to assess the relationship of the prescribing of inhaled steroids and short-acting β2-agonists to patient and physician characteristics. Patient characteristics included race-ethnicity, insurance status, age, sex, geographic region, metropolitan area status, and patient visit status. The patient's medical insurance status was classified as private (Blue Cross/Blue Shield or other private or commercial insurance) and nonprivate (Medicare, Medicaid, workers' compensation, other insurance, and self-pay). Patient age was dichotomized to younger than 18 years (children) and 18 years and older (adults). A new patient visit was defined as a patient being seen by the physician for the first time or if the purpose of the visit was for a general medical examination. For the purposes of this study, physician specialty was categorized as generalists (ie, general-family practitioners, internists, and pediatricians), asthma specialists (ie, allergists and pulmonologists), and all others. Physician practice settings included private offices (NAMCS) and hospital outpatient departments (NHAMCS). Among the selected characteristics, patient race-ethnicity and physician specialty are of particular interest. Changes in racial-ethnic and specialty differences over time were examined by defining interaction terms in logistic regression models; however, no significant difference was found. Overall, inhaled steroids were reportedly prescribed during 24% (99% CI, 21% to 26%) and short-acting β2-agonists during 55% (95% CI, 52% to 58%) of all asthma visits. Several characteristics significantly altered the odds of physician-reported prescribing of inhaled steroids and short-acting β2-agonists (Table I). After controlling for other characteristics, the likelihood of visits by Hispanics being prescribed inhaled steroids was 43% less than that in visits by whites, whereas patients cared for by asthma specialists were 3 times more likely to be prescribed inhaled steroids than those seen by generalists. Physician-reported prescribing of short-acting β2-agonists did not differ by patient race-ethnicity or physician specialty. Compared with their adult counterparts, children and adolescent visitors were 43% less likely to be prescribed inhaled steroids but 80% more likely to be prescribed short-acting β2-agonists. Relative to visits to hospital outpatient departments, visitors to in-office physicians were less likely to be prescribed both inhaled steroids (66% lower) and short-acting β2-agonists (61% lower). In addition, asthma visits by patients with private insurance versus those without private insurance had 1.51 times the odds of being prescribed short-acting β2-agonists. Physician-reported prescribing did not vary significantly on the basis of the remaining characteristics. The success of asthma treatment hinges on appropriate medication use according to evidence-based practice guidelines, such as those published by NAEPP.2National Asthma Education Program (NAEP) Expert panel report: guidelines for the diagnosis and management of asthma update on selected topics—2002.J Allergy Clin Immunol. 2002; 110: 141-219Abstract Full Text PDF Google Scholar, 3National Asthma Education Program (NAEP) Expert panel report II: guidelines for the diagnosis and management of asthma. National Heart, Lung, and Blood Institute, Bethesda (MD)1997Google Scholar It is evident from this and many past studies that appropriate use of drugs in these 2 dominant treatment classes remains elusive. Many studies have attempted to identify underlying reasons for the continuing discrepancies between current asthma care and evidence-based guidelines. Our study reveals that physician nonadherence to NAEPP guidelines in the use of inhaled steroids and short-acting β2-agonists is not uncommon during ambulatory asthma visits in the United States. Moreover, asthma pharmacotherapy in children, minorities, patients covered by nonprivate insurance, and generalists' patients conforms less well to consensus guidelines. These disparities in asthma care do not appear to have decreased in recent years. Our data preclude the determination of reasons for the observed prescribing variations, but some speculations can be made on the basis of published studies.2National Asthma Education Program (NAEP) Expert panel report: guidelines for the diagnosis and management of asthma update on selected topics—2002.J Allergy Clin Immunol. 2002; 110: 141-219Abstract Full Text PDF Google Scholar, 7Chambers CV Markson L Diamond JJ Lasch L Berger M Health beliefs and compliance with inhaled corticosteroids by asthmatic patients in primary care practices.Respir Med. 1999; 93: 88-94Abstract Full Text PDF PubMed Scopus (164) Google Scholar, 8Vollmer WM O'Hollaren M Ettinger KM Stibolt T Wilkins J Buist AS et al.Specialty differences in the management of asthma. A cross-sectional assessment of allergists' patients and generalists' patients in a large HMO.Arch Intern Med. 1997; 157: 1201-1208Crossref PubMed Google Scholar Apparently less optimal asthma treatment in children might be attributed to concerns about potential adverse effects of inhaled steroids, family-maternal factors, and child-parental health beliefs regarding asthma and the value of prevention. For racial-ethnic minorities, there might be multiple barriers, including access to health care, patient-clinician communication, and cultural differences in health beliefs. Variations in practice among physicians might be due to differences in training and knowledge of the disease and relevant guidelines, as well as in organizational structure and resource allocation. Regardless of the underlying reasons, the observed disparities manifest a practical departure from guidelines, which recommend no differential therapy for any subpopulation other than different dose, number, and frequency of medications by asthma severity.2National Asthma Education Program (NAEP) Expert panel report: guidelines for the diagnosis and management of asthma update on selected topics—2002.J Allergy Clin Immunol. 2002; 110: 141-219Abstract Full Text PDF Google Scholar, 3National Asthma Education Program (NAEP) Expert panel report II: guidelines for the diagnosis and management of asthma. National Heart, Lung, and Blood Institute, Bethesda (MD)1997Google Scholar Addressing these disparities is particularly important in that asthma morbidity can be prevented and use of health care resources can be reduced by appropriate drug treatment. The current findings should be interpreted in the context of several limitations of NAMCS and NHAMCS data. Both data sources are serial cross-sectional surveys of patient visits recorded by health care providers. An inherent limitation of any survey study is response bias; also, reported medication use only reflects a physician's best knowledge of any new and continued prescriptions. In addition, lack of patient-level clinical data limits our assessment of the appropriateness of medication regimens because of the inability to adjust for some known correlates of asthma medication use: drug dosage, asthma severity, symptom frequency, duration and frequency of health care use, and patient compliance in filling and using prescribed medications. Nonetheless, we have no reason to believe that the above data limitations differentially affect subgroups of the study sample according to the selected characteristics so as to compromise the validity of cross-group comparisons. By using data from physician-based national surveys, this study suggests that the prescribing of 2 dominant classes of asthma medications (ie, inhaled steroids and short-acting β2-agonists) varies by nonclinical factors. These factors pinpoint subpopulations of patients with asthma whose drug regimens need to be reassessed and groups of physicians whose practices for treating asthma need to be improved. These findings should be used to guide future interventions that aim to bring physician practice into agreement with published guidelines for optimal asthma treatment.