Trends in Outpatient Visits with Benzodiazepines among US Adults With and Without Bronchitis or Chronic Obstructive Pulmonary Disease from 1999 to 2010

Abstract

Little is known about trends in prescriptions for benzodiazepines among patients with chronic obstructive pulmonary disease (COPD). Our objective was to examine trends of office/outpatient department visits with a mention of a benzodiazepine made by patients aged ≥40 years with COPD in the United States. We used data from the National Ambulatory Medical Care Survey and National Hospital Ambulatory Medical Care Survey from 1999–2010. From 1999 to 2010, the estimated numbers of office/outpatient department visits with a benzodiazepine mentioned increased from 20.7 million to 43.2 million among all patients, from 684,000 to 1.5 million among patients with COPD, and from 20.0 million to 41.7 million among patients without COPD. Using all 12-years of data, patients with COPD were more likely to have a visit with a mention of a benzodiazepine than patients without COPD (adjusted prevalence ratio = 1.48, 95% CI = 1.27–1.71).The unadjusted percentage of all office/outpatient department visits by patients with COPD with a mention of a benzodiazepine increased from 4.6% during 1999–2002 to 10.2% during 2007–2010 (P trend < 0.001). After adjustment for age, sex, and race, the adjusted prevalence ratio for 2007–2010 compared with 1999–2002 was 2.26 (95% confidence interval: 1.60–3.17). Since 1999, the number and percentage of office/outpatient department visits with a mention of a benzodiazepine by patients with COPD and all patients may have increased in the United States.

Keywords:

• benzodiazepines
• COPD
• prescriptions
• therapeutics
• trends
• United States

Background

Chronic obstructive pulmonary disease (COPD) remains a challenging ­condition in the United States and is a growing concern across the globe (1). In the United States, an estimated 6.3% of adults aged ≥18 years or about 15 million adults reported ever having been told by a health professional that they have COPD (2), and approximately 14% of adults have spirometric ­evidence of obstructive lung function (3). Worldwide, 210 million adults worldwide are estimated to have COPD (4). As a result of their disease, people with COPD commonly have depression and anxiety (5). Furthermore, a large percentage of people with COPD also experience insomnia (6).

Benzodiazepines are often used in the pharmacologic treatment of depression, anxiety, and insomnia. Because concerns about the effect of benzodiazepines on respiration in patients with COPD go back decades, authors have urged caution in the use of these agents in patients with COPD. Consequently, a recent analysis of administrative data in Canada showing that approximately 32% of patients aged ≥66 years with COPD had received a new benzodiazepine was surprising (7). Furthermore, two studies of patients with COPD who were prescribed benzodiazepines noted increased complications and mortality (8, 9).

Because little is known about the use of benzodiazepines by patients with COPD in the United States, our objective was to examine trends in office/outpatient department visits with a mention of a benzodiazepine among these patients. Our secondary objective was to compare these trends among office/outpatient department visits made by patients with COPD to trends among office/outpatient department visits made by patients without COPD and all patients.

Methods

We used nationally representative data from the National Ambulatory Medical Care Survey (NAMCS) and National Hospital Ambulatory Medical Care Survey (NHAMCS) from 1999 to 2010 that were collected by the Centers for Disease Control and Prevention, National Center for Health Statistics (10). NAMCS is an annual survey of a national probability sample of ambulatory visits to non-federally employed physicians engaged in office-based, patient care as classified by the American Medical Association or the American Osteopathic Association. Visits to Community Health Centers were added in 2006. NAMCS uses a multistage probability design that includes the following stages of ­selection: primary sampling units (counties, county equivalents (such as parishes and independent cities), towns, townships, minor civil divisions), physician practices within primary sampling units, and patient visits within practices. The physician-patient encounter or visit is the basic sampling unit in NAMCS. Physician response rates ranged from 58.3% in 2010 to 70.4% in 2002. Institutional Review Board review was not required.

The NHAMCS is a national probability sample survey of visits to hospital outpatient and emergency departments. The sampling frame includes noninstitutional general and short-stay hospitals in the 50 states and the District of Columbia, but not Federal, military, and Veterans Administration hospitals. Sampling occurs at four stages of selection: primary sampling units, hospitals within primary sampling units, clinics within hospitals, and patient visits within clinics over a randomly assigned 4-week reporting period. Like the NAMCS the basic sampling unit is the patient visit. Only outpatient visits were included in the present study. Hospital response rates ranged from 89.4% in 2007 to 96.0% in 2000.

Physicians, their staff, or U.S. Census Bureau field representatives performed data collection and completed patient record forms. Visits listing International Classification of Diseases, 9th revision, Clinical Modification (ICD-9-CM) codes 490 (bronchitis not specified as acute or chronic), 491 (chronic bronchitis), 492 (emphysema), or 496 (chronic airway obstruction, not elsewhere classified, which includes COPD) on any of three diagnosis fields were considered to have been made by a patient with COPD. Up to six medications could be listed from 1999–2002 and up to eight medications from 2003-2010. We searched for the following benzodiazepines: alprazolam, chlordiazepoxide, clobazam, clonazepam, clorazepate, diazepam, estazolam, flurazepam, lorazepam, midazolam, nitrazepam, oxazepam, temazepam, and triazolam. Because a listing of a benzodiazepine in the medication fields of the data bases could indicate that the benzodiazepine was prescribed, provided, or continued, we will refer to mentions of benzodiazepines (11).

We included age, gender, and race as covariates. Information concerning race was based on the physician's knowledge of the patient or on the physician's or assistant's judgment rather than patient self-report. Because information about race was missing from substantial percentages of records, we used the NCHS imputed data for race (whites, blacks, and other).

The analyses were limited to office/outpatient department visits made by patients who were aged ≥40 years to limit the inclusion of patients who might have had asthma. Using combined data for NAMCS and NHAMCS, we present results for the numbers and percentage of all office/outpatient department visits that listed COPD as a physician's diagnosis that also included a benzodiazepine in the patient's medication record. Although the patient visit is the fundamental sampling unit, we also present estimates of benzodiazepine mentions among patients by using a methodology to calculate patient-level estimates (12). Because only up to six medications could be recorded from 1999 to 2002 rather than eight medications from 2003 to 2010, analyses covering the period from 1999 to 2010 were based on using the first six listed medications. Tests for trend were conducted by using orthogonal polynomial contrasts. Log-linear regression analysis was used to estimate prevalence ratios for predictors of visits with a mention of a benzodiazepine. Statistical analyses using sampling weights were conducted with SAS-callable SUDAAN to account for the complex survey design.

Results

Table 1 shows the annual unweighted sample sizes and estimated weighted numbers of outpatient department visits and physician office visits with and without a diagnosis of COPD for patients aged ≥40 years. In any given year, COPD was reported for about 3% of all NHAMCS outpatient department visits and NAMCS physician office visits. The combined NAMCS and NHAMCS estimated weighted numbers of office/outpatient department visits made by patients with COPD ranged from 12,991,000 in 2000 to 22,493,000 in 2002. For office/­outpatient department visits made by patients with COPD, no significant linear trends by year were observed for mean age (range 64.3–67.8 years; P trend = 0.087), the percentage of office visits made by men (range 40.0%–50.5%, P trend = 0.534), and the percentage of office ­visits made by white patients (range 84.0%–95.3%, P trend = 0.152).

Table 1. Number of unweighted and weighted office/outpatient department office visits among patients aged ≥40 years, National Ambulatory Medical Care Survey and National Hospital Ambulatory Medical Care Survey

	1999	2000		2001		2002		2003		2004		2005		2006		2007		2008		2009		2010
Total visits
Unweighted numbers
NHAMCS-OPD	13,310	13,168		14,853		16,335		16,361		15,242		14,501		16,107		17,370		17,269		17,035		18,129
NAMCS	12,956	16,838		15,473		17,530		15,871		16,455		16,392		18,086		20,783		17,777		20,625		19,396
Weighted numbers
NHAMCS-OPD	39,614,000	41,131,000		40,777,000		39,281,000		43,444,000		41,063,000		43,122,000		52,209,000		45,813,000		56,763,000		50,295,000		53,015,000
NAMCS	449,325,000	476,216,000		528,670,000		525,055,000		543,250,000		555,645,000		595,812,000		539,832,000		599,586,000		594,969,000		655,892,000		614,543,000
Total	488,939,000	517,347,000		569,446,000		564,336,000		586,695,000		596,708,000		638,934,000		592,042,000		645,399,000		651,732,000		706,187,000		667,557,000
COPD visits
Unweighted numbers
NHAMCS-OPD	512	382		315		342		305		319		321		292		279		407		366		348
NAMCS	304	277		293		456		299		333		333		455		394		360		442		405
Weighted numbers
NHAMCS-OPD	1,362,000	1,123,000		1,020,000		1,134,000		1,206,000		1,106,000		1,424,000		1,379,000		1,182,000		1,860,000		1,294,000		1,201,000
NAMCS	14,783,000	11,867,000		17,199,000		21,358,000		15,279,000		16,395,000		19,411,000		18,686,000		14,983,000		15,110,000		19,132,000		18,478,000
Total	16,145,000	12,991,000		18,219,000		22,493,000		16,486,000		17,501,000		20,835,000		20,065,000		16,165,000		16,970,000		20,426,000		19,679,000
Non-COPD visits
Unweighted numbers
NHAMCS-OPD	12,798	12,786		14,538		15,993		16,056		14,923		14,180		15,815		17,091		16,862		16,669		17,781
NAMCS	12,652	16,561		15,180		17,074		15,572		16,122		16,059		17,631		20,389		17,417		20,183		18,991
Weighted numbers
NHAMCS-OPD	38,252,000	40,007,000		39,757,000		38,146,000		42,238,000		39,957,000		41,698,000		50,830,000		44,631,000		54,903,000		49,001,000		51,814,000
NAMCS	434,542,000	464,349,000		511,471,000		503,697,000		527,971,000		539,250,000		576,400,000		521,146,000		584,603,000		579,859,000		636,760,000		596,065,000
Total	472,794,000	504,356,000		551,228,000		541,843,000		570,209,000		579,207,000		618,099,000		571,976,000		629,234,000		634,763,000		685,761,000		647,879,000

COPD, chronic obstructive pulmonary disease; NAMCS, National Ambulatory Medical Care Survey; NHAMCS-OPD, National Hospital Ambulatory Medical Care Survey-Outpatient Department.

Results based on outpatient visits from 1999 to 2010

From 1999 to 2010, the estimated annual numbers of office/outpatient department visits with a mention of a benzodiazepine increased from 20.7 million to 43.2 million (2.1-fold increase) among all patients, from 684,000 to 1.5 million (2.2-fold increase) among patients with COPD, and from 20.0 million to 41.7 ­million (2.1-fold increase) among patients without COPD (Figure 1).

Figure 1. Average annual numbers of office/outpatient department office visits and patients with a mention of a benzodiazepine, National Ambulatory Medical Care Survey and National Hospital Ambulatory Medical Care Survey. Results are shown for the numbers of visits made by patients with chronic obstructive pulmonary disease (Panel A), the numbers of visits made by patients without chronic obstructive pulmonary disease (Panel B), the numbers of patients with chronic obstructive pulmonary disease (Panel C), and the numbers of patients without chronic obstructive pulmonary disease (Panel D).

Among patients with COPD, the percentage of office/outpatient department visits with a mention of benzodiazepine increased from 4.6% (SE 0.6) during 1999-2003 to 10.2% (SE 1.1) during 2007–2010 (P trend < 0.001) (Table 2). Significant increases in percentages of visits by patients with COPD with a mention of a benzodiazepine were observed among men and women, Whites, and all age groups.

Table 2. Percentage of office/outpatient visits and patients with a mention of a benzodiazepine, by chronic obstructive pulmonary disease status, National Ambulatory Medical Care Survey and National Hospital Ambulatory Medical Care Survey

	Visits				Patients
	1999-2002	2003-2006	2007-2010	P trend	2001-2002	2003-2006	2007-2010	P trend
Total
Total-unadjusted	4.3 (0.2)	4.8 (0.2)	6.4 (0.2)	<0.001	3.0 (0.2)	3.7 (0.2)	4.7 (0.2)	<0.001
Total-age-adjusted	4.3 (0.2)	5.0 (0.2)	6.7 (0.2)	<0.001	3.0 (0.2)	3.8 (0.2)	4.7 (0.2)	<0.001
Age (years)
40-49	4.6 (0.3)	5.5 (0.3)	7.5 (0.3)	<0.001	2.9 (0.3)	3.8 (0.2)	4.6 (0.3)	<0.001
50-59	4.6 (0.3)	5.5 (0.3)	6.9 (0.3)	<0.001	3.2 (0.3)	4.3 (0.3)	5.1 (0.4)	<0.001
60-69	4.0 (0.3)	4.4 (0.3)	6.1 (0.3)	<0.001	2.7 (0.3)	3.6 (0.3)	4.7 (0.4)	<0.001
≥70	3.9 (0.2)	4.0 (0.2)	5.5 (0.2)	<0.001	3.1 (0.3)	3.2 (0.2)	4.5 (0.3)	0.001
Gender
Women	4.8 (0.2)	5.6 (0.2)	7.6 (0.3)	<0.001	3.3 (0.2)	4.2 (0.2)	5.4 (0.3)	<0.001
Men	3.7 (0.2)	4.1 (0.2)	5.4 (0.2)	<0.001	2.5 (0.2)	3.1 (0.2)	3.8 (0.3)	<0.001
Race
White	4.5 (0.2)	5.4 (0.2)	7.1 (0.2)	<0.001	3.0 (0.2)	4.0 (0.2)	5.0 (0.3)	<0.001
Black	3.5 (0.5)	3.1 (0.3)	4.9 (0.3)	0.001	2.4 (0.5)	2.8 (0.4)	3.3 (0.4)	0.058
Other	3.2 (0.7)	1.9 (0.3)	3.7 (0.5)	0.341	—*	1.5 (0.3)	2.4 (0.4)	0.827
COPD visits
Total-unadjusted	4.6 (0.6)	6.8 (1.1)	10.2 (1.1)	<0.001	2.9 (0.6)	5.1 (0.9)	9.5 (1.5)	<0.001
Total-age-adjusted	3.9 (0.7)	6.8 (1.4)	9.9 (1.2)	<0.001	2.3 (0.5)	4.8 (1.0)	7.8 (1.3)	<0.001
Age (years)
40-49	—*	—*	9.8 (2.4)	0.034	—*	—*	—*	—
50-59	—*	—*	9.5 (1.8)	0.001	—*	—*	—*	—
60-69	4.1 (1.2)	6.8 (1.7)	9.7 (2.1)	0.021	—*	—*	9.9 (2.9)	0.009
≥70	5.9 (1.1)	5.2 (1.2)	11.0 (1.7)	0.014	4.7 (1.3)	4.0 (1.0)	12.7 (2.9)	0.012
Gender
Women	4.4 (1.1)	7.4 (1.7)	12.5 (1.7)	<0.001	—*	5.8 (1.5)	10.0 (2.0)	<0.001
Men	3.4 (0.7)	6.1 (1.6)	6.5 (1.3)	0.004	—*	3.6 (0.9)	—*	0.018
Race
White	3.9 (0.8)	7.4 (1.6)	10.5 (1.3)	<0.001	2.3 (0.6)	5.2 (1.1)	8.0 (1.3)	<0.001
Black	—*	—*	—*	—	—*	—*	—*	—
Other	—*	—*	—*	—	—*	—*	—*	—
No COPD visits
Total-unadjusted	4.2 (0.2)	4.7 (0.2)	6.3 (0.2)	<0.001	3.0 (0.2)	3.7 (0.2)	4.6 (0.2)	<0.001
Total-age-adjusted	4.3 (0.2)	4.9 (0.2)	6.6 (0.2)	<0.001	3.0 (0.2)	3.7 (0.2)	4.6 (0.2)	<0.001
Age (years)
40-49	4.6 (0.3)	5.5 (0.3)	7.4 (0.3)	<0.001	3.0 (0.3)	3.8 (0.2)	4.6 (0.3)	<0.001
50-59	4.6 (0.3)	5.4 (0.3)	6.9 (0.3)	<0.001	3.2 (0.3)	4.2 (0.3)	5.0 (0.4)	0.001
60-69	4.0 (0.3)	4.3 (0.3)	6.0 (0.3)	<0.001	2.7 (0.3)	3.6 (0.3)	4.6 (0.4)	<0.001
≥70	3.8 (0.2)	3.9 (0.2)	5.2 (0.2)	<0.001	3.0 (0.3)	3.1 (0.2)	4.2 (0.3)	0.002
Gender
Women	4.8 (0.2)	5.5 (0.2)	7.4 (0.3)	<0.001	3.3 (0.2)	4.1 (0.2)	5.2 (0.3)	<0.001
Men	3.7 (0.2)	4.1 (0.2)	5.3 (0.2)	<0.001	2.5 (0.2)	3.1 (0.2)	3.8 (0.3)	<0.001
Race
White	4.5 (0.2)	5.3 (0.2)	7.0 (0.2)	<0.001	3.0 (0.2)	4.0 (0.2)	5.0 (0.3)	<0.001
Black	3.4 (0.5)	3.1 (0.3)	4.8 (0.3)	0.002	2.4 (0.5)	2.8 (0.4)	3.3 (0.4)	0.075
Other	3.2 (0.7)	1.9 (0.3)	3.7 (0.5)	0.332	—*	1.5 (0.3)	2.5 (0.4)	0.828

COPD, chronic obstructive pulmonary disease.

*Estimates do not meet standards of precision or reliability (relative standard errors >30%).

Using data for the entire 12-year study period, the percentage of visits with mention of a benzodiazepine among all office/outpatient department visits was lower in patients aged ≥60 years than those aged 40–49 years, was higher among women than men, was higher among whites than patients of other races, was higher among patients with COPD than those without COPD, and increased during the study period after adjustment for covariates (Table 3). Generally, the patterns of office/outpatient department visits with a mention of a benzodiazepine pertained to patients with COPD as well, but the confidence intervals were wide because of the smaller sample sizes. Although the temporal increase in office/outpatient department with a mention of a benzodiazepine appeared stronger among patients with COPD (adjusted prevalence ratio [PR] = 2.26, 95% confidence interval [CI] = 1.60, 3.17) than without COPD (adjusted PR = 1.51, 95% CI = 1.37, 1.66), the interaction term between COPD status and time period did not achieve statistical significance (P = 0.068).

Table 3. Adjusted* prevalence ratios (95% confidence interval) for office/outpatient department visits and patients with a mention of a benzodiazepine (yes/no), National Ambulatory Medical Care Survey and National Hospital Ambulatory Medical Care Survey, 1999–2010

	Visits			Patients
Age (years)	Total	COPD	No COPD	Total	COPD	No COPD
40-49	1.00	1.00	1.00	1.00	1.00	1.00
50-59	0.97 (0.91, 1.04)	1.36 (0.85, 2.18)	0.96 (0.90, 1.03)	1.11 (1.00, 1.23)	2.06 (1.00, 4.24)	1.10 (0.99, 1.22)
60-69	0.83 (0.77, 0.89)	1.19 (0.74, 1.92)	0.82 (0.76, 0.89)	0.98 (0.88, 1.10)	2.16 (1.03, 4.53)	0.97 (0.86, 1.08)
≥70	0.74 (0.69, 0.80)	1.28 (0.82, 1.98)	0.73 (0.68, 0.79)	0.92 (0.83, 1.03)	2.58 (1.32, 5.07)	0.90 (0.81, 1.00)
Gender
Men	1.00	1.00	1.00	1.00	1.00	1.00
Women	1.42 (1.35, 1.50)	1.62 (1.25, 2.11)	1.41 (1.34, 1.49)	1.41 (1.29, 1.54)	2.19 (1.45, 3.29)	1.38 (1.27, 1.51)
Race
White	1.00	1.00	1.00	1.00	1.00	1.00
Black	0.69 (0.62, 0.76)	0.60 (0.35, 1.04)	0.69 (0.62, 0.77)	0.69 (0.59, 0.82)	0.56 (0.22, 1.39)	0.70 (0.59, 0.82)
Other	0.51 (0.42, 0.61)	0.17 (0.04, 0.71)	0.52 (0.43, 0.62)	0.57 (0.39, 0.83)	0.14 (0.03, 0.61)	0.58 (0.40, 0.84)
COPD status
Yes	1.48 (1.27, 1.71)	—	—	1.55 (1.25, 1.93)	—	—
No	1.00	—	—	1.00	—	—
Time period
1999-2002	1.00	1.00	1.00	1.00^†	1.00^†	1.00^†
2003-2006	1.13 (1.02, 1.26)	1.48 (0.96, 2.26)	1.12 (1.01, 1.24)	1.25 (1.08, 1.45)	1.67 (0.97, 2.89)	1.24 (1.06, 1.44)
2007-2010	1.53 (1.39, 1.68)	2.26 (1.60, 3.17)	1.51 (1.37, 1.66)	1.60 (1.37, 1.88)	3.21 (1.91, 5.39)	1.56 (1.33, 1.84)

COPD, chronic obstructive pulmonary disease.

*Adjusted for other variables shown in Table.

†Reference period is 2001–2002.

To provide comparable data to publications of authors who examined benzodiazepine use among elderly patients with COPD, we analyzed data for office/outpatient department visits made by patients aged ≥65 years with COPD. The percentage of office/outpatient department visits with a mention of a benzodiazepine increased from 5.0% during 1999-2002 to 10.7% during 2007–2010 (P trend < 0.001). This represented an approximately 2–fold increase (unadjusted PR = 2.12, 95% CI = 1.38, 3.26) (Figure 2).

Figure 2. Unadjusted prevalence ratio (95% confidence interval) for office/outpatient department visits with a mention of a benzodiazepine (yes/no) among patients with chronic obstructive pulmonary disease aged ≥ 65 years, National Ambulatory Medical Care Survey and National Hospital Ambulatory Medical Care Survey.

Results based on patients from 2001 to 2010

The annual number of patients with COPD and a ­mention of a benzodiazepine increased from 183,000 in 2001 to 655,000 in 2010. The unadjusted percentage of patients with COPD and a mention of a benzodiazepine increased from 2.9% to 9.5% (P trend < 0.001) during the study period (Table 2), and the relative increase among patients with COPD (239%) exceeded that among patients without COPD (54%). After adjustment for age, sex, race, and time period, patients with COPD were 55% more likely to have a mention of a benzodiazepine than patients without COPD (Table 3). The interaction between COPD status and time was significant (P = 0.027) suggesting that the increase in the percentage of patients with COPD and a mention of a benzodiazepine was stronger than the increase among patients without COPD.

Discussion

Our analyses of nationally-representative data of office/outpatient department visits show that the number of visits with a mention of a benzodiazepine among patients with and without COPD more than doubled from 1999 to 2010 in the United States. Furthermore, patients with COPD were more likely to have an office/outpatient department visit with a mention of a benzodiazepine in the medical record.

Because many patients with COPD suffer from insomnia (6), anxiety, and depression (5), they are potential candidates for treatment with benzodiazepines. Concerns about the effects of benzodiazepine use on respiration harken back decades, and these concerns are especially pertinent to patients with impaired lung function such as COPD. One of the earliest investigations in this area found small effects of intravenous diazepam administration on respiratory parameters in patients with COPD (13). This was followed by additional case reports that raised concerns about the effects of several benzodiazepines on aspects of respiration in patients with respiratory failure and chronic bronchitis (14, 15, 16, 17). An early trial of 6 patients with exacerbations of their COPD found that nitrazepam administration led to significant decreases in FEV1 and FVC and nonsignificant changes in paO₂ and paCO₂ (18). A subsequent trial of two benzodiazepines found decreases in minute ventilation and tidal volume, an increase in respiratory frequency, and respiratory acidosis (19). Another study noted central depression of respiration in patients with chronic bronchitis who were treated with a benzodiazepine (20). Effects of benzodiazepines on PO₂ (21–24), PCO₂ (21–24), and CO₂ sensitivity (25) have also been observed among patients with COPD. However, other studies have not shown adverse effects of benzodiazepines on respiratory parameters in patients with COPD (26–29).

Although authors have cautioned about the use of ­benzodiazepines in patients with COPD, major professional organizations have either not addressed the issue in their guidelines or have done so only in a limited manner. The American Thoracic Society/European Respiratory Society does not recommend the use of hypnotics, which include benzodiazepines, for insomnia in patients with COPD (30), and the Canadian Thoracic Society does not recommend using benzodiazepines to treat dyspnea (31). Recent guidelines from the Global Initiative for Chronic Obstructive Lung Disease guidelines and the American College of Physicians, American College of Chest Physicians, American Thoracic Society, and European Respiratory Society do not address the appropriate use of ­benzodiazepines in patients with COPD.

The research examining the possible effects of benzodiazepines on respiration in patients with COPD has recently charted a new direction by examining the impact that benzodiazepine use has on complications and mortality in patients with COPD (8, 9). In a study of 177,355 Canadian patients aged ≥66 years with COPD, patients who were new users of benzodiazepines were more likely to experience respiratory exacerbations (relative risk [RR] = 1.45, 95% CI: 1.36, 1.54) and emergency room visits for COPD or pneumonia than nonusers of benzodiazepines (RR = 1.92, 95% CI: 1.69, 2.18) (8). No significant differences for hospitalizations for COPD or pneumonia or admissions to the intensive care unit during hospitalizations for COPD or pneumonia were observed. Interestingly, mortality was slightly lower among new users than nonusers of benzodiazepines (RR = 0.91, 95% CI: 0.85, 0.98). A Swedish study of 2249 patients aged ≥ 45 years with COPD who were started on oxygen therapy found no difference in rates of hospital admissions between patients who used benzodiazepines and those who did not (9). However, mortality during a median follow-up of 1.1 years was elevated among users of benzodiazepines (adjusted hazard ratio 1.21, 95% confidence interval 1.05, 1.39). Although these two studies have raised new concerns about the use of benzodiazepines in patients with COPD, additional studies will be required to examine the effects of a gamut of benzodiazepines on a range of potential adverse outcomes across a range of COPD severity.

The prevalence of use of benzodiazepines by patients with COPD is poorly understood. Consequently, the estimate that 31% of Canadian patients aged ≥66 years used a benzodiazepine was surprising (7). We estimate that a benzodiazepine was mentioned on 10.7% of office visits by and for 12.2% of patients aged ≥65 years with COPD during 2007–2010.

The study results should be viewed in the context of several limitations. The small numbers of patients with COPD necessitated aggregating several years of data to produce stable estimates and precluded detailed comparisons among demographic subgroups. Second, COPD severity could not be assessed. Use of benzodiazepines among patients with the most severe forms of COPD is of great interest as these patients are potentially at most risk for the respiratory depressing effects of benzodiazepines. Third, visits by patients with COPD were identified using ICD-9-CM codes in one of three diagnoses codes. If COPD was not among these codes, some visits by patients with COPD might not be identified in the study sample. To understand the possible effect of not capturing such patients with COPD on the results, knowledge of benzodiazepine use patterns by these patients would be needed and such information is generally not available. If benzodiazepine use among such patients tended to resemble that of patients without COPD, then the estimated percentages of visits with a benzodiazepine prescription and the slope of the increase were likely overestimated. Fourth, medication dosages were not available. Finally, information about socioeconomic status was not available. Because some but not all studies indicate that the prevalence of benzodiazepine use tends to be inversely associated with educational status (32, 33), examining benzodiazepine use by educational status or other socioeconomic indicators would have been informative considering that little is known about the socioeconomic associations with benzodiazepine use among patients with COPD. Furthermore, including socioeconomic indicators in the regression models might have changed some of the prevalence ratios, particularly that of COPD status given that socioeconomic status is also inversely related to COPD status.

In conclusion, our analyses suggest that office/­outpatient department visits with a mention of a benzodiazepine have increased among patients aged ≥40 years with COPD. Because benzodiazepine use has been linked to adverse outcomes, the increase in the mention of these medications in our analyses raises questions about the reasons for the increase in these medications. Patients with COPD are at a double disadvantage: they are at increased risk for common adverse outcomes of benzodiazepine use like members of the general population and, because of their impaired respiratory function, they may be more susceptible to respiratory side effects of these medications than adults with normal respiratory function. Additional investigations into the use of benzodiazepines by patients with COPD seem warranted.

Disclaimer

The findings and conclusions in this article are those of the authors and do not necessarily represent the official position of the Centers for Disease Control and ­Prevention.

Declaration of Interest Statement

Neither of the authors has a conflict of interest. The authors alone are responsible for the content and ­writing of the paper.