Effectiveness of Bed Bug Monitors for Detecting and Trapping
Bed Bugs in Apartments
CHANGLU WANG,1WAN-TIEN TSAI,2RICHARD COOPER,1,3AND JEFFREY WHITE3
J. Econ. Entomol. 104(1): 274Ð278 (2011); DOI: 10.1603/EC10141
Bed bugs, Cimex lectularius L., are now considered a serious urban pest in the United
States. Because they are small and difÞcult to Þnd, there has been strong interest in developing and
using monitoring tools to detect bed bugs and evaluate the results of bed bug control efforts. Several
bed bug monitoring devices were developed recently, but their effectiveness is unknown. We
comparatively evaluated three active monitors that contain attractants: CDC3000, NightWatch, and
a home-made dry ice trap. The Climbup Insect Interceptor, a passive monitor (without attractants),
was used for estimating the bed bug numbers before and after placing active monitors. The results of
the Interceptors also were compared with the results of the active monitors. In occupied apartments,
ice trap (operated for 1 d) and trapped more bed bugs than CDC3000 and NightWatch (operated for
numbers of bed bugs. CDC3000 and the dry ice trap operated for 1 d were equally as effective as the
designed to be able to operate for several consecutive nights. When operated for four nights,
and evaluating the results of bed bug control programs.
bed bug, Cimex lectularius, monitoring, traps, dry ice
Bed bug, Cimex lectularius L., infestations are increas-
ing rapidly on a global scale, and bed bug resurgence
will probably last for many years, in part, due to lack
of effective detection and control techniques (Ter
Poorten and Prose 2005, Harlan 2006, Kilpinen et al.
may spread rapidly through passive and active dis-
persal of bed bugs (Wang et al. 2010). Low-income
communities are more likely to suffer chronic and
increased bed bug infestations in the years to come
effective community-wide eradication of infestations.
are not always aware that an infestation exists for a
variety of reasons. Some people do not react to bed
bed bug bites with poison ivy, mosquito bites, and
other nonÐbed bug-related causes, allowing the bugs
difÞcult to Þnd, particularly when just a few bugs are
present (Wang et al. 2009a,b). They rest in very se-
cretive hiding places or areas that may not be readily
accessible for inspection. Visual inspections are labor
intensive and are often unreliable (Pinto et al. 2008).
Interviews with residents are not reliable either be-
cause many people do not recognize the presence of
bed bug infestations or are unwilling to report the
problem (Wang et al. 2010).
Using dogs trained to detect bed bugs has been a
low level infestations. It is a very efÞcient detection
method. However, this method has several disadvan-
tages: 1) it is not readily available in all parts of the
and 3) the accuracy can vary signiÞcantly from one
dogÐhandler team to another (Pinto et al. 2008).
Lack of affordable and effective tools for detecting
early infestation stages. In turn, it leads to greater
difÞculty and increased costs associated with eradi-
cation efforts. In response to the need for affordable
and effective bed bug monitoring tools, both passive
(without attractants) and active (with attractants)
monitors have been developed. In a recent study, the
1Corresponding author: Department of Entomology, Rutgers Uni-
versity, New Brunswick, NJ 08901 (e-mail: firstname.lastname@example.org.
2Current address: Department of Entomology, National Taiwan
University, Taipei 106, Taiwan.
3Cooper Pest Solutions, Lawrenceville, NJ 08648.
0022-0493/11/0274Ð0278$04.00/0 ? 2011 Entomological Society of America
proved to be more effective than visual inspections in
detecting bed bugs (Wang et al. 2009a). Based on the
same principles, a commercial product (Climbup In-
sect Interceptor, Susan McKnight, Inc., Memphis,
TN), was subsequently developed and proved to be
effective in detecting bed bugs (Wang et al. 2009b).
However, Interceptors cannot be used under furni-
designed to monitor rooms that are vacant.
with CO2(CO2cylinder as source), chemical lures,
and heat. They found the trap was able to catch large
numbers of bed bugs in a vacant apartment. Wang et
(dry ice as source) and chemical lure can detect the
presence of bed bugs that are not detected by visual
inspections. This was the Þrst study showing that an
inexpensive active monitor can be used in detecting
bed bugs and assessing the effectiveness of bed bug
Two commercial monitors, CDC3000 (Cimex Sci-
To date, there are no reported studies on the effec-
tiveness of bed bug active monitors in Þeld settings.
Examining the effectiveness of bed bug monitors un-
der Þeld conditions is necessary to provide accurate
information on the potential role of these monitoring
tools in bed bug management. The objective of this
study was to determine 1) the effectiveness of active
bed bug monitors compared with visual inspection in
effectiveness of three active bed bug monitors. The
three active monitors were CDC3000, NightWatch,
and a home-made dry ice trap (baited with dry ice).
Materials and Methods
Study Sites. The study sites were two high-rise
apartment buildings located in Bayonne, NJ. All units
in each building were one-bedroom or studio apart-
ments of ?36Ð45 m2. One or two elderly people re-
pyrethroid sprays at least one month before the study
as part of the regular pest management contract. Dia-
tomaceous earth dust residues were visible on ßoors
and furniture in all apartments included in this study.
Bed Bug Monitors. The CDC3000 monitors were
monitors were purchased from the manufacturer.
Both monitors use CO2, heat, and synthetic lures to
their concentrations were not disclosed by the man-
after referred as dry ice trap) was made from an
inverted cat feeder covered with white polyester fab-
size of the cat feeder was 35.5 by 17.5 by 7 cm (length
by width by height; Van Ness Plastics, Clifton, NJ).
The fabric size was same as the outer surfaces of the
cat feeder using 1.27-cm-wide masking tape. The in-
terior surface of the inverted cat feeder was coated
bugs from escaping. A 1,127-ml insulated jug
ice pellets was placed on top of the cat feeder as CO2
CO2to escape into the atmosphere. Unlike the
and lure, the dry ice trap only uses CO2to attract bed
Each CDC3000 used a 166-g CO2cylinder contain-
ing 46 g of liquid CO2. Once screwed onto the
CDC3000 trap, the cylinder continuously released
CO2at ?42 ml/min for 10 h. Each NightWatch used
programmed by the manufacturer to release CO2for
8 h each day. The median CO2release rate (deter-
mined by measuring the average CO2cylinder weight
loss per night based on measurement from 10 Night-
Watch monitors) was 161 ml/min. At this rate, each
The dry ice trap released CO2at a rate of 731Ð801
ml/min for 12?15 h in a room maintained at 23Ð25?C.
times the CO2emitted by an adult human (Leff and
Climbup Insect Interceptors (hereafter referred to
of selecting test apartments, comparing it with active
monitors, and conÞrming the presenceÐabsence of
bed bugs after deploying the active monitors. The
Interceptors were provided by Susan McKnight, Inc.
Experiment I. Three heavily infested apartments
located on three ßoors were identiÞed to compara-
itors. Brief visual inspections found at least 426 bed
bugs (adults and nymphs) in each apartment. Imme-
and an insulated jug.
A dry ice trap consisted of an inverted cat feeder
February 2011WANG ET AL.: EFFECTIVENESS OF BED BUG MONITORS
The monitors were placed around the infested area
(sofa, bed, or piles of clothing) in each apartment
between 1 p.m. and 6 p.m. The monitors were exam-
ined the next day (?24 h later). Each monitor was
then placed in a different apartment on each of the
next 2 d following a Latin square design. On each day,
the monitors were examined and reset in the after-
noon. Bed bug counts from two dry ice traps and one
CDC3000 monitor were estimated nearest to the 10Õs
programmed to release CO2immediately after setting
up on each day. Thus, all monitors released CO2im-
mediately after setup on each day.
Experiment II. Fifteen lightly infested apartments
located on eight ßoors were used to evaluate the
small numbers of bed bugs. To select the apartments,
we Þrst installed Interceptors under beds, sofas, or
in 39 apartments with reported infestations. In most
apartments, bridges (e.g., bed linens touching the
ßoor) between the furniture and the ßoors or walls
were not completely removed due to lack of resident
cooperation or large footing of the bed frames. These
conditions might have negatively affected the effec-
tiveness of the Interceptors. After 7 d of deployment,
the Interceptors were removed and examined for the
number of bed bugs. We (C.W., R.C., J.W.) immedi-
ately conducted a visual inspection of the apartment.
The mean inspection time was 13 min per apartment
by three people (total, 39 min per apartment). The
Fifteen apartments with one to 35 bed bugs in total
collected from Interceptors plus visual inspection
NightWatch, and Þve dry ice traps were randomly
trapped bed bugs was recorded the next day. The
experiment was continued for another 2 d. On each
day, the three types of monitors were rotated among
type of monitor during the 3-d period. NightWatch
monitors were programmed to release CO2immedi-
ately after installation.
Experiment III. This experiment was to evaluate
the daily trapping pattern of NightWatch in lightly
infested apartments and determine the effectiveness
tiple days. Eight of the original 15 apartments in Ex-
The eight apartments were chosen by placing Inter-
ceptors in the 15 apartments for 10 d and selecting
units that had four to 25 trapped bugs. The Intercep-
tors were deployed for 10 d instead of 7 d as in
experiment II because of scheduling difÞculties. Im-
mediately after examining and removing the Inter-
ceptors, a NightWatch was placed beside the infested
furniture (bed or sofa) in each apartment. All Night-
Watch monitors used 567-g CO2cylinders that al-
lowed the monitors to operate for four consecutive
nights. The monitors were programmed to begin re-
leasing CO2at 20:00 h each day as per manufacturerÕs
instructions. The NightWatch monitors were exam-
ined daily for four consecutive days. The trapped bed
bugs were counted and then removed. All CO2cyl-
inders attached to the monitors still contained CO2at
the end of the 4-d experiment. After removing the
once again placed under the legs of infested furniture
conÞrm the presenceÐabsence of bed bugs after de-
ployment of NightWatch monitors.
Data Analysis. Analysis of variance (ANOVA) was
conducted to compare the effectiveness of the mon-
itors using Proc Mixed (for bed bug counts) or Proc
Genmod (for bed bug infestation detection rate) in
SAS software (SAS Institute 2003). The bed bug
counts in experiment I were logarithmic transformed
II were square root transformed before ANOVA.
Means were separated using the StudentÕs least sig-
niÞcant difference separation test.
Experiment I. In heavily infested apartments, an
average of 850 ? 390, 207 ? 99, and 58 ? 28 bed bugs
was trapped overnight by dry ice trap, CDC3000, and
NightWatch, respectively. Relative effectiveness of
the three monitors was as follows: dry ice trap ?
CDC3000 ? NightWatch (F ? 269.3, df ? 2, 2; P ?
dry ice trap, CDC3000, and NightWatch per day was
1,365, 329, and 95, respectively.
Experiment II. The effectiveness of bed bug mon-
ments were divided into two groups: very low level
infestations (?10 bed bugs per apartment) and low
level infestations (11Ð35 bugs per apartment). Night-
Watch was signiÞcantly less effective than the other
16.3, df ? 4, P ? 0.003). In apartments with low level
Table 1. Comparative effectiveness of bed bug monitors for detecting bed bug infestations in apartments
Initial count of bed bugs per
apartment from Interceptors
and visual inspections
Detection rate (% detected infestations)
CDC3000 Night Watch
Note: The Interceptors were installed under furniture legs 7 d before visual inspections. Other monitors were deployed for 1 d.
276JOURNAL OF ECONOMIC ENTOMOLOGY
Vol. 104, no. 1
infestations, all monitors were able to detect ?80% of
Among the 15 lightly infested apartments, an aver-
age of 4.7 ? 1.5, 1.3 ? 0.4, and 1.1 ? 0.8 bed bugs was
trapped overnight by dry ice trap, CDC3000, and
The dry ice trap caught signiÞcantly more bed bugs
apartment from Interceptors (7-d deployment) be-
fore and after deployment of monitors were 7.3 ? 2.5
and 5.2 ? 1.5, respectively. The mean count from
visual inspections was 2.5 ? 1.4. The Interceptors
operated for 7 d trapped similar number of bed bugs
as the dry ice trap and trapped more bed bugs than
CDC3000 and NightWatch operated for 1 d. Based on
daily counts from Interceptors (total counts divided
by the number of deployment days), the Interceptors
were as effective as visual inspection, CDC3000, and
NightWatch and were less effective than dry ice trap
(F ? 2.6; df ? 5, 70; P ? 0.04).
Experiment III. The mean counts per apartment
from Interceptors before and after deployment of
NightWatch were 12.5 ? 2.3 and 5.3 ? 2.1, respec-
on the Þrst day and an additional apartment on the
second day. It failed to trap bed bugs from one apart-
1.3 ? 0.8, 1.1 ? 0.6, and 0.8 ? 0.4 from day 1 to day 4,
4.6 ? 2.1 per apartment. It was not signiÞcantly dif-
ferent from the Interceptor counts (original counts
adjusted by a factor of 4/10 to reßect the differences
in trapping period) (F ? 1.47; df ? 2, 19; P ? 0.26).
The study revealed that dry ice trap, CDC3000,
NightWatch, and the Climbup Interceptor are effec-
tive monitors in detecting the presence of small num-
bers of bed bugs. When operated for just 1 d, dry ice
traps and CDC3000 were equally effective as visual
inspections for detecting very low level infestations.
Placing and examining monitors require less skill and
experience as visual inspections. In apartment build-
suspected having bed bug infestations, deployment of
aiding in the detection of bed bug infestations.
The effectiveness of the monitors varied signiÞ-
cantly. On a daily basis, NightWatch was the least
effective monitor among the active monitors, but its
effectiveness can be enhanced by running the device
ufacturer. Perhaps the most signiÞcant discovery is
that the simple and inexpensive home-made dry ice
trap is more effective than the two expensive com-
mercial active monitors (?$450 for NightWatch and
$1,000 for CDC3000). The higher effectiveness and
affordability make dry ice traps very appealing to
those on a limited budget. The major disadvantage of
using dry ice trap is the potential risk of injury from
accidental ingestion or exposure to dry ice. When
properly secured and operated, dry ice traps can be a
very valuable monitoring tool in bed bug manage-
Wang et al. (2009c) showed CO2was the most
effective attractant to bed bugs compared with heat
and chemical lure. All of the three active monitors
tested use CO2. However, their CO2release rates
varied greatly. Dry ice trap released CO2at ?4 times
as that by NightWatch and nearly 20 times of that by
CDC3000. Unlike CDC3000 and NightWatch, dry ice
trap only employs CO2as attractant. Thus, the differ-
ential CO2rates may have at least partially accounted
for the higher effectiveness of dry ice trap compared
with CDC3000 and NightWatch.
Bed bugs take a bloodmeal every week or more
we may assume that only a small proportion of bed
bugs in an occupied room searches for hosts each
established infestations. Therefore, one should never
assume zero catch implies the room being monitored
is bed bug free. The multiple-night trapping with
NightWatch veriÞed that multiple night deployment
one night deployment.
This study was conducted in apartments where hu-
man hosts were present before and during the de-
ployment of active monitors. The presence of human
host might have competed with the active monitors
and negatively affected the bed bug counts from the
monitors. Our preliminary data show that hungry bed
bugs are much more responsive to active monitors
unoccupied for a few days or longer, effectiveness of
This study further conÞrms that the passive Inter-
ceptors are an effective monitoring tool for detecting
very low number of bed bugs (Wang et al. 2009a,b).
presence of human host in the bed turns the passive
unt per monitor
ean bed bug cou
Dry ice trapCDC3000 NightWatch
itors (1-d deployment). Different letters above the bars in-
dicate signiÞcant differences (P ? 0.05; ANOVA).
Bed bug counts (mean ? SEM) from three mon-
February 2011WANG ET AL.: EFFECTIVENESS OF BED BUG MONITORS
Interceptor into an active monitor. When placed for Download full-text
7 d, Interceptors can be equally effective as 1-d de-
ployment of dry ice traps and are a safe and cost
effective method for detecting low numbers of bed
bugs. It should be noted that the effectiveness of
Interceptors was compromised in most apartments
or ßoors (e.g., bed linens touching the ßoor). In ad-
dition, Interceptors were not installed under all up-
holstered furniture in each apartment.
Failure to detect bed bugs in their early stage in-
festation is recognized as a major challenge by pest
management professionals. With the recent develop-
ment of both active and passive monitors, greater
control effectiveness may be achieved through more
efÞcient early detections. Monitors may have the po-
tential to reduce the need for pesticide applications
because they can remove the few bugs that survive
treatments. Additional studies on the effect of trap
designs and environmental conditions (i.e., occupied
effect of trapping on bed bug populations will be
We are grateful to John Wilson and Boyd Gonnerman for
Þeld assistance; Susan McKnight, Inc., for providing Inter-
ceptors; Bayonne Housing Authority for providing study
sites; and Cimex Science for providing CDC3000. This
project is sponsored by Cooper Pest Solutions and U.S. De-
partment of Urban and Housing Development Healthy
Homes Technical Studies grant program. This is New Jersey
Experiment Station publication D-08-08117-04-10.
Anderson, J. F., F. J. Ferrandino, S. McKnight, J. Nolen, and
J.Miller. 2009. Acarbondioxide,heatandchemicallure
Harlan, H. 2006. Bed bugs 101: the basics of Cimex lectu-
larius. Am. Entomol. 52: 99Ð101.
Kilpinen, O., K.-M.V. Jensen, and M. Kristensen. 2008. Bed
bug problems in Denmark, with a European perspective,
pp. 395Ð399. In W. H. Robinson and D. Bajomi (eds.),
Proceedings of the 6th International Conference on Ur-
ban Pests, 13Ð16 July 2008, Budapest, Hungary.
Leff, A. R., and P. T. Schumacker. 1993. Respiratory phys-
iology basics and application. W. B. Saunders Co., Phil-
Pinto, L. J., R. Cooper, and S. K. Kraft. 2008. Bed bug hand-
bookÑthe complete guide to bed bugs and their control.
Pinto & Associates, Inc. Mechanicsville, MD.
Potter, M. F. 2008. The business of bed bugs. 2008. Pest
Manag. Profess. 76: 24Ð25, 28Ð32, 34, 36Ð40.
Reinhardt, K., D. Isaac, and R. Naylor. 2010. Estimating the
feeding rate of the bed bug Cimex lectularius in an in-
fested room: an inexpensive method and a case study.
Med. Vet. Entomol. 24: 46Ð54.
2009. Sensitivity to bites by the bedbug, Cimex lectu-
larius. Med. Vet. Entomol. 23: 163Ð166.
SAS Institute. 2003. SAS/STAT userÕs guide, version 9.1,
SAS Institute, Cary, NC.
Ter Poorten, M. C., and N. S. Prose. 2005. The return of the
common bedbug. Pediatr. Dermatol. 22: 183Ð187.
Usinger, R. L. 1966. Monograph of Cimicidae (Hemiptera:
Heteroptera). Thomas Say Foundation, College Park,
Wang,C.,T.Gibb,andG.W.Bennett. 2009a. Evaluationof
two least toxic integrated pest management programs for
managing bed bugs (Heteroptera: Cimicidae) with dis-
Wang, C., T. J. Gibb., and G. W. Bennett. 2009b. Intercep-
tors assist in bed bug monitoring. Pest Control Technol.
37: 112, 114.
Wang, C., T. J. Gibb., G. W. Bennett, and S. McKnight.
2009c. Bed bug attraction to pitfall traps baited with
Wang, C., T. J. Gibb., and G. W. Bennett. 2010. Character-
istics of Cimex lectularius (Hemiptera: Cimicidae), infes-
tation and dispersal in a high-rise apartment building. J.
Econ. Entomol. 103: 172Ð177.
Received 15 April 2010; accepted 12 October 2010.
278JOURNAL OF ECONOMIC ENTOMOLOGY
Vol. 104, no. 1