Journal of Urban Health

, Volume 91, Issue 6, pp 1076–1086

Bedbug Complaints among Public Housing Residents—New York City, 2010–2011

Authors

    • Arctic Investigations ProgramCenters for Disease Control and Prevention
  • Nancy Ralph
    • Epidemiology and Biostatistics ProgramCity University of New York School of Public Health at Hunter College
  • Andrew Maroko
    • Department of Earth, Environmental, and Geospatial SciencesCity University of New York School of Public Health at Lehman College
  • Lorna Thorpe
    • Epidemiology and Biostatistics ProgramCity University of New York School of Public Health at Hunter College
Article

DOI: 10.1007/s11524-013-9859-y

Cite this article as:
Gounder, P., Ralph, N., Maroko, A. et al. J Urban Health (2014) 91: 1076. doi:10.1007/s11524-013-9859-y

Abstract

Few studies have evaluated population-level risk factors for having a bedbug infestation. We describe characteristics associated with bedbug complaints among New York City Housing Authority (NYCHA) residents. Unique households receiving bedbug extermination services in response to a complaint during January 1, 2010 to December 31, 2011 were identified from NYCHA’s central facilities work order database. We examined associations between household characteristics and having a bedbug complaint using a generalized estimating equation Poisson regression model, accounting for clustering by housing development. Of the 176,327 NYCHA households, 11,660 (6.6 %) registered a bedbug complaint during 2010–2011. Bedbug complaints were independently associated with households having five or more children versus no children (prevalence ratio [PR] = 2.0), five or more adults versus one adult (PR = 1.6), a head of household (HOH) with impaired mobility (PR = 1.3), a household member receiving public assistance (PR = 1.2), a household income below poverty level (PR = 1.1), and a female HOH (PR = 1.1). Infestations were less likely to be reported by households with employed members (PR = 0.9), and an HOH aged 30–44 years (PR = 0.9) or 45–61 years (PR = 0.9), compared with an HOH aged 18–29 years. These results indicate that bedbug control efforts in public housing should be targeted toward households with low income and high occupancy.

Keywords

BedbugsPublic housingEpidemiologyEnvironmental health

Introduction

Pest control professionals in major cities were among the first to notice the resurgence of bedbugs in the late 1990s.13 These pests are now of increasing concern to residents in cities across the USA.1,46 In the 2009 New York City (NYC) Community Health Survey (CHS), a random-digit-dialed telephone survey of adult noninstitutionalized residents, 7 % reported a problem with bedbugs that required an exterminator during the previous year.7 Although bedbugs have not been implicated as vectors for disease transmission, they can cause dermatologic symptoms and be harmful to psychological health.4,811 Bedbugs can affect all socioeconomic groups and result in adverse economic consequences, including extermination costs, and damaged furniture.3,5,12,13

The ecology and life cycle of bedbugs (most commonly CIMEX lectularius) make eradication difficult. They hide in tight spaces during the day and come out at night to feed. Bedbugs depend entirely on a blood meal from a host for nutrition; yet they can survive 6–12 months without feeding. During their life cycle, bedbugs can reproduce in large numbers.4,14 Eradication often requires that pest management professionals employ either traditional pest management practices involving pesticide application or a multifaceted approach known as integrated pest management (IPM), using nonchemical approaches (e.g., improved monitoring, and sanitation).4,15

Epidemiologic research on bedbugs has been limited, and systems to monitor infestations at a population level do not exist.4,16 This lack of research activity or tracking is likely caused in part by bedbugs not being a substantial concern during the latter half of the 20th century.1,2 In light of the recent resurgence of bedbugs, however, identifying risk factors and health effects associated with having bedbug infestations is now necessary for devising effective bedbug control strategies.4,17

The New York City Housing Authority (NYCHA) is the largest public housing authority in the USA with a mission of serving low- and middle-income residents. NYCHA operates 2,596 residential buildings, clustered into 334 developments, and houses 5 % of NYC’s residents.18,19 NYCHA maintains a centralized database on all apartment-related complaints from tenants, including bedbug complaints. The availability of data on bedbug complaints from a defined population offered a unique opportunity for understanding bedbug epidemiology. In this study, we describe characteristics associated with bedbug complaints among NYC public housing residents.

Methods

Data Source for Study Population

Information on characteristics of public housing residents was obtained from the NYCHA Tenant Data System (TDS). Data contained in the TDS include age, sex, race/ethnicity, family composition (number of adults and children), and income by income source. The TDS data used for this study represent the NYCHA tenant population as of January 1, 2011. We defined all residents living in a single apartment unit as a household. As a measure of crowding, NYCHA classifies housing units as overoccupied if there are twice as many household members as the number of living and bedrooms.20 NYCHA classifies persons as disabled if they self-report disability/impairment for vision, hearing, mental, or mobility and as senior if aged ≥62 years. For this study, we restricted our definition of disabled persons to those with impaired mobility.

Identifying Unique Households with a Probable Bedbug Infestation

All work orders referencing a complaint of bedbugs during January 1, 2010 to December 31, 2011 were abstracted from NYCHA’s administrative database. Because the presence of bedbugs at the time of inspection by building maintenance staff was not routinely documented, bedbug complaints that resulted in a work order for extermination services were used as a proxy for true infestations. For analytic purposes, a household was defined as having a probable bedbug infestation if it had one or more work orders for bedbug extermination during the study period. A household might have required multiple follow-up exterminations to resolve an infestation. Work orders that referenced public spaces (e.g., stairwells and hallways), that were canceled before any work was started, or that had no demographic information available for the household members were excluded from analysis. Because households might have received a follow-up extermination service for a complaint that originated outside the study period or received extermination service for recurrence of a prior infestation, the data reflect prevalent, not incident, infestations.

Statistical Methods

Prevalence of probable bedbug infestations among NYCHA households was calculated by head of household and household characteristics for the 2-year timeframe of January 1, 2010 to December 31, 2011. Poisson regression was used to calculate crude and adjusted prevalence ratios (PRs), and 95 % confidence intervals (CIs). Effect modification of age on the association between disability and probable infestation was assessed; the interaction term was not statistically significant and was excluded from the final model. Excess correlation between the number of household members and the number of bedrooms in a unit and between variables indicating socioeconomic status was evaluated. Only household size and number of bedrooms were determined to have high correlation (Pearson correlation coefficient = 0.80; p < 0.001). Therefore, the final multivariate model excluded the number of bedrooms. Generalized estimating equations were used to account for clustering of multiple infested households in the same NYCHA development. These statistical calculations were performed by using SAS® version 9.2 (SAS Institute, Inc. Cary, North Carolina, USA).

ArcGIS (Esri, Redlands, California, USA) was used to depict the prevalence of probable bedbug infestations among NYCHA households by development and self-reported bedbug prevalence, according to the 2009 CHS telephone survey data by zip code-aggregated neighborhoods as defined by the United Hospital Fund (UHF).7,21 Because the CHS results comprise a single year, the map depicts probable bedbug infestations among NYCHA households identified only during 2010. For ease of visualization on maps, all prevalence estimates were categorized into color-coded quartiles. The correlation between the burden of infestations within specific NYCHA developments and the surrounding neighborhood’s burden was assessed by calculating a Spearman correlation.

Results

Study Population

There were 176,327 households in the TDS as of January 1, 2011 (Table 1). Three quarters of households were headed by females, 80 % of households were headed by persons aged 30–74 years, and 38 % of households were headed by persons with a disability. The majority of households were headed by non-Hispanic black (47 %) and Hispanic (43 %) persons. Half of the households had family incomes below the US federal poverty level. The majority of households had no children aged <18 years.
TABLE 1

Prevalence of probable bedbug infestations among New York City Housing Authority households, 2010–2011

Characteristics

Households with ≥1 bedbug complaint, No. (%)

All households, No. (%)

Prevalence of bedbug complaints (%)

Total

11,660

176,327

6.6

Head of household characteristics

 Age (years)

  18–29

809 (7)

10,587 (6)

7.6

  30–44

2,925 (25)

39,880 (23)

7.3

  45–61

4,028 (35)

62,150 (35)

6.5

  62–74

2,163 (19)

38,564 (22)

5.6

  ≥75

1,735 (15)

25,133 (14)

6.9

  Missinga

0

13 (0)

0

 Sex

  Male

2,256 (19)

135,548 (77)

6.9

  Female

9,404 (81)

40,779 (23)

5.5

 Race/ethnicity

  Non-Hispanic Asian

225 (2)

7,598 (4)

3.0

  Non-Hispanic black

6,463 (55)

82,168 (47)

7.9

  Hispanic

4,422 (38)

76,438 (43)

5.8

  Other

78 (1)

1,343 (1)

5.8

  Non-Hispanic white

472 (4)

8,780 (5)

5.4

 Impaired mobility?

  No

7,204 (62)

116,617 (66)

6.2

  Yes

4,456 (38)

59,710 (34)

7.5

Socioeconomic status

 Below federal poverty line

  No

5,020 (43)

85,277 (48)

5.9

  Yes

6,507 (56)

88,966 (50)

7.3

  Missing

133 (1)

2,084 (1)

6.4

 Family receiving public assistance

  No

9,773 (84)

156,299 (89)

6.3

  Yes

1,887 (16)

20,028 (11)

9.4

 ≥1 employed household member

  No

6,658 (57)

93,041 (53)

7.2

  Yes

5,002 (43)

83,286 (47)

6.0

Household characteristics

 Number of bedrooms in housing unit

  0

420 (4)

6,721 (4)

6.2

  1

2,383 (20)

39,124 (22)

6.1

  2

5,235 (45)

83,914 (48)

6.2

  3

2,976 (25)

39,797 (23)

7.5

  ≥4

646 (6)

6,771 (4)

9.5

 Number of adults in householdb

  1

6,059 (52)

96,487 (55)

6.3

  2

3,725 (32)

55,627 (32)

6.7

  3

1,346 (12)

17,580 (10)

7.7

  4

401 (3)

5,139 (3)

7.8

  ≥5

129 (1)

1,494 (1)

8.6

 Number of children in householdc

  0

6,580 (56)

112,549 (64)

5.8

  1

2,354 (20)

32,158 (18)

7.3

  2

1,533 (13)

19,017 (11)

8.1

  3

733 (6)

8,301 (5)

8.8

  4

297 (3)

2,897 (2)

10.3

  ≥5

163 (1)

1,405 (1)

11.6

 ≥2 household members for every living room and bedroom

  No

11,046 (95)

169,959 (96)

6.5

  Yes

614 (5)

6,368 (4)

9.6

Probable bedbug infestation defined as a work order for extermination services in response to a bedbug complaint

aIndicated only for variables with missing observations

bAdults aged ≥18 years

cChildren aged <18 years

Prevalence of Probable Bedbug Infestation

We identified 31,245 noncancelled work orders for bedbug complaints registered by 13,496 unique households during 2010–2011. A total of 11,660 unique households received one or more treatments for bedbugs and met our definition of having a probable bedbug infestation, yielding a 2-year prevalence for probable bedbug infestation among NYCHA households of 6.6 %. The number of exterminations per household ranged from 1 to 13, with an average of 1.7. The majority (52 %) of households received a single extermination treatment.

Factors Associated with Probable Bedbug Infestation

The multivariable analysis demonstrated a significant association between having a probable bedbug infestation and living in households headed by a female (PR = 1.1; 95 % CI 1.0–1.2) or someone with a disability (PR = 1.3; 95 % CI 1.2–1.4) (Table 2). Compared with households with a non-Hispanic white head of household, households headed by Asians were less likely to have a probable infestation (PR = 0.6; 95 % CI 0.5–0.7) and households headed by non-Hispanic blacks were more likely to have an infestation (PR = 1.4; 95 % CI 1.2–1.6). Households headed by persons aged 30–44 years (PR = 0.9; 95 % CI 0.8–1.0) or 45–61 years (PR = 0.9; 95 % CI: 0.8–1.0) were at a reduced risk for having a probable bedbug infestation, compared with those aged 18–29 years.
TABLE 2

Association between characteristics of New York City Housing Authority households and probable bedbug infestations, 2010–2011

 

Crude PR

95 % CI

Adjusted PR

95 % CI

Head of household characteristic

 Age (years)

  18–29

Ref

 

Ref

 

  30–44

1.0

0.9–1.0

0.9*

0.8–1.0

  45–61

0.8

0.8–0.9

0.9*

0.8–1.0

  62–74

0.7

0.7–0.8

0.8

0.7–0.9

  ≥75

0.9

0.8–1.0

1.0

0.9–1.2

 Female

1.3

1.2–1.3

1.1*

1.0–1.2

 Race/ethnicity

  Non-Hispanic Asian

0.6

0.5–0.7

0.6

0.5–0.7

  Non-Hispanic black

1.5

1.3–1.7

1.4

1.2–1.6

  Hispanic

1.1

0.9–1.2

1.0

0.9–1.2

  Other

1.1

0.8–1.4

1.0

0.8–1.3

  Non-Hispanic white

Ref

 

Ref

 

 Impaired mobility

1.2

1.2–1.3

1.3

1.2–1.4

Socioeconomic status characteristic

  Below federal poverty line

1.2

1.2–1.3

1.1

1.1–1.2

  Family receiving public assistance

1.5

1.4–1.6

1.2

1.1–1.2

  ≥1 household member employed

0.8

0.8–0.9

0.9

0.8–0.9

Household characteristic

 Number of bedrooms in housing unit

  0

Ref

 

 

  1

1.0

1.0–1.1

 

  2

1.2

1.1–1.3

 

  3

1.2

1.1–1.4

 

  ≥4

1.4

1.2–1.6

 

 Number of adults in householda

  1

Ref

 

Ref

 

  2

1.1*

1.0–1.1

1.2

1.1–1.3

  3

1.2

1.1–1.3

1.4

1.3–1.5

  4

1.2

1.1–1.4

1.5

1.3–1.6

  ≥5

1.4

1.2–1.6

1.6

1.3–1.9

 Number of children in householdb

  0

Ref

 

Ref

 

  1

1.3

1.2–1.3

1.3

1.2–1.3

  2

1.4

1.3–1.5

1.4

1.3–1.5

  3

1.5

1.4–1.6

1.6

1.4–1.7

  4

1.8

1.6–2.0

1.8

1.6–2.0

  ≥5

2.0

1.7–2.3

2.0

1.7–2.4

 ≥2 household members for every living room and bedroom

1.5

1.4–1.6

0.9*

0.8–1.0

CI confidence interval, PR prevalence ratio

Probable bedbug infestations defined as a work order for a bedbug complaint that resulted in extermination services

*p < 0.05

aAdults aged ≥18 years

bChildren aged <18 years

For NYCHA residents, households of lower socioeconomic status were more likely to have a probable bedbug infestation than higher income households. Households with an income below the US poverty level (PR = 1.1; 95 % CI 1.1–1.2) or receiving public assistance (PR = 1.2; 95 % CI 1.1–1.2) had higher prevalence of probable infestation. Households in which one or more members were employed were less likely to have a probable infestation (PR = 0.9; 95 % CI 0.8–0.9).

The prevalence of probable bedbug infestations directly correlated with the number of children, the number of adult household members, and the number of bedrooms in the housing unit. Households with five or more adults were more likely to have a probable infestation, compared with households with one adult (PR = 1.6; 95 % CI 1.3–1.9). Households with five or more children were more likely to have a probable infestation than households without children (PR = 2.0; 95 % CI 1.7–2.4). Housing units with four or more bedrooms were more likely to have a probable infestation than units with zero bedrooms (PR = 1.4; 95 % CI 1.2–1.6). Living in overoccupied households was associated with a lower prevalence of having a probable bedbug infestation (PR = 0.9; 95 % CI 0.8–1.0).

The prevalence of probable bedbug infestation among NYCHA households grouped by development during 2010 and the self-reported prevalence of bedbugs by UHF, according to the 2009 CHS results, were overlaid onto a single map (Fig. 1). The location and size of each development are depicted on the map. The majority of NYCHA developments appear to cluster in northern Manhattan, southern Bronx, and central Queens. The majority of NYCHA households in the top quartile of probable bedbug infestations, both by development and UHF neighborhood, were located in northern Manhattan. According to the CHS results, northern Manhattan also falls in the top quartile of self-reported bedbug prevalence.
https://static-content.springer.com/image/art%3A10.1007%2Fs11524-013-9859-y/MediaObjects/11524_2013_9859_Fig1_HTML.gif
FIGURE 1.

Prevalence of bedbug complaints among New York City Housing Authority (NYCHA) developments and by neighborhood, 2010*. *Boundaries for 34 zip code-aggregated New York City (NYC) neighborhoods as defined by the United Hospital Fund. Data are classified by quartiles. Percentage of unique households identified from NYCHA administrative database as receiving extermination services during 2010 for a bedbug complaint. §Percentage of households reporting a problem with bedbugs requiring an exterminator during the previous year, according to the 2009 Community Health Survey, a random-digit-dialed telephone survey of NYC residents.

Prevalence of probable bedbug infestation among NYCHA households grouped by UHF neighborhood during 2010 was moderately correlated with the surrounding neighborhood’s self-reported bedbug prevalence, according to the 2009 CHS survey (Spearman correlation coefficient 0.56; p < 0.01). Among all 176,327 NYCHA households, 6,132 (3.5 %) had a probable infestation in 2010 and 11,660 (6.6 %) had a probable infestation during 2010–2011.

Discussion

This study of bedbug complaints among NYCHA residential households during 2010–2011 identified that large families, households of low socioeconomic status, and households headed by persons with impaired mobility were at higher risk for having a probable bedbug infestation. The prevalence of bedbug infestations among NYCHA households is less than the citywide self-reported bedbug prevalence in the 2009 CHS and, on the basis of this study’s results, is the greatest in housing developments located in northern Manhattan.

Plausible mechanisms exist for which the risk for bedbug infestation would be increased by household size and poverty. For example, households with fewer members might be protected because bedbugs are known to travel between locations by riding on persons or their belongings.4,14 Therefore, having fewer household members visiting potentially infested venues might reduce the total risk for acquiring bedbugs. Bedbugs are known to disperse across adjacent dwellings in multi-unit buildings.22,23 Thus, the association between household size and infestations might be the consequence of larger households living in larger apartment units with correspondingly greater exposure to potentially infested units (adjacent, above, or below) than households in smaller housing units. Being of lower socioeconomic status might be associated with certain practices that introduce bedbugs into homes (e.g., acquiring infested second-hand furniture).12 After the infestation is established, households without the financial means to afford bedbug control measures (e.g., purchasing mattress encasements and frequent laundering) might complicate eradication and prolong the infestation.22

The reasons for the observed associations between sex, race, crowding, and having a probable bedbug infestation among NYCHA households are unclear. Although race and sex are often correlated with poverty, the population in our study is restricted to low- and moderate-income households and the multivariable model was adjusted for multiple indicators of socioeconomic status. Therefore, this association between race/ethnicity, sex, and probable infestation likely represents confounding from variables not included in the model. For instance, CHS revealed that prevalence of self-reported bedbug infestations varies between different UHF neighborhoods. If non-Hispanic blacks or females who are the head of the household were more likely to live in developments located in high-prevalence neighborhoods, the association might be the result of their greater environmental exposure to bedbugs. Another unexpected study finding was that overoccupied (i.e., crowded) housing units were less likely to have a probable bedbug infestation after adjusting for the total number of adults and for total number of children in the household. Although crowding has been believed to increase the risk for bedbug infestations,12 the absolute number of persons in the housing unit might be a more important factor.

The prevalence of probable bedbug infestations among NYCHA households in 2010 (3.5 %) was lower than the self-reported prevalence among the general NYC population in the 2009 CHS (6.7 %).7 The results should be comparable between our study and CHS because both studies rely on self-reported bedbug infestations that resulted in extermination services, even though the data sources used to estimate the prevalence are different. NYCHA has certain unique advantages that can result in a lower prevalence of bedbug complaints by their residents. First, the risk for acquiring bedbugs increases with rapid turnover of tenants in an apartment building because bedbugs can be transported on bedding, clothing, and furniture.24 NYCHA buildings typically have lower turnover than other NYC buildings; the average tenure of an NYCHA household is 20 years (NYCHA unpublished data, 2012). In addition, NYCHA provides extermination at no charge to tenants, which might result in infestations being reported and resolved earlier than in buildings where tenants are responsible for that cost. Eradicating bedbugs in a single unit without simultaneously treating infestations in surrounding units is difficult;12,25 therefore, NYCHA building maintenance staff might inspect neighboring units when they suspect an infestation has spread. NYCHA has also trained its staff to employ IPM techniques. Although not specifically studied for bedbugs, IPM has been reported to be more effective than the traditional pest management in eliminating other pests (e.g., cockroaches and rodents) in NYCHA apartments.15 Fewer barriers to reporting possible infestations, building-level coordination of extermination, and use of IPM by NYCHA can all influence reduction of risk for infestation recurrence after successful eradication.

The study has certain limitations. Relying on data collected for administrative purposes rather than research required making certain assumptions in defining a probable bedbug infestation. For instance, visible confirmation of the presence of bedbugs was not routinely recorded and exterminations might have occurred in households with no bedbugs. Using exterminations as a proxy for true infestation might have overestimated the prevalence of infestation. Conversely, it is possible certain households had infestations that were unreported and would not have received extermination services, which would have resulted in an underestimate of the prevalence. The extent to which these limitations offset each other cannot be quantified; therefore, the prevalence estimates should be interpreted with caution. Further, the demographic profile of tenants likely differs from the general NYC population because of NYCHA’s distinct mission of providing decent, affordable housing and the asymmetric geographic distribution of its developments. Therefore, the results are not generalizable beyond NYC public housing.

This study is among the first to formally describe the epidemiology of bedbugs within a defined population. The results can direct public health and other officials in bedbug control efforts. Consideration should be given to actively monitor for infestations among populations at high risk (e.g., apartment buildings with high occupancy). Similarly, certain populations might benefit from targeted education regarding steps that can be taken to reduce the risk for acquiring a bedbug infestation (e.g., seal cracks and crevices in walls, avoid bringing discarded furniture into the home). In parallel with our study, NYCHA maintenance personnel were trained to identify and reliably document the presence of bedbugs in the work order database. Documentation of the presence of verified bedbugs, not just complaints, might improve future monitoring of bedbugs in NYCHA buildings. A prospective case–control study in which housing units with a verified bedbug infestation are enrolled and matched with housing units that do not have an infestation can evaluate health effects, economic consequences, and other potential risk factors for bedbugs by collecting information unavailable through administrative data systems. Health departments might consider developing surveillance systems to better monitor bedbug infestations and associated health effects. Lastly, social-network analysis (SNA) can complement traditional epidemiologic tools used for understanding risk factors for acquiring bedbugs. SNA methods quantitatively describe associations between persons and places and have helped elucidate the person-to-person spread of other conditions (e.g., obesity and smoking behavior).2628 SNA might be similarly helpful in understanding the spread of bedbugs because they are transported between places by persons and their belongings.

Acknowledgments

The authors thank Calcedonio Bruno, Anthony Porcelli, and Luis Ponce for their insights into the pest management and tenant complaint management procedures at NYCHA. We acknowledge Tamara Dumanovksy and Anne-Marie Flatley for abstracting the data for analysis from NYCHA’s administrative database and providing feedback on the manuscript. We acknowledge Sharon Balter, Marci Layton, and Julie Magri for their guidance in the design of the study and reviewing drafts of the manuscript.

Note

The findings and conclusions in this report are those of the author(s) and do not necessarily represent the official position of the Centers for Disease Control and Prevention or the New York City Housing Authority.

Funding source

New York Public Housing Authority and City University of New York School of Public Health at Hunter College

Copyright information

© The New York Academy of Medicine (outside the USA) 2014