Table 2: Selected Characteristics of Samples, by city--United States, National HIV Behavioral Surveillance System: Injecting Drug Users, May 2005-February 2006

Metropolitan Statistical Area (“City”) IDU Population Sizea NHBS-IDU Sample Sizeb Sampling Fractionc Proportion Eligibled Proportion Recruited by a Stranger e Cross-Recruitment by Field Site Cross-Recruitment by Race/ Ethnicity g
No. No. %  % % % %
Atlanta, Georgia 14,602 616 4.2 91 12 18 17
Baltimore, Maryland 58,720 785 1.3 92 20 21 25
Boston, Massachusetts 67,044 540 0.8 88 2 30 35
Chicago, Illinois 32,206 653 2.0 83 4 46 18
Dallas, Texas 31,931 620 1.9 92 3 35 27
Denver, Colorado 20,689 612 3.0 87 4 74 43
Detroit, Michigan 27,166 568 2.1 96 3 n/af 16
Fort Lauderdale, Florida 7,375 441 6.0 87 8 36 32
Houston, Texas 34,117 662 1.9 90 1 n/af 32
Las Vegas, Nevada 13,708 341 2.5 98 17 8 40
Los Angeles, California 98,616 661 0.7 91 2 7 42
Miami, Florida 9,280 740 8.0 82 3 25 34
Nassau, New York 12,177 557 4.6 95 0.4 0.2 47
New Haven, Connecticut 13,629 593 4.4 90 2 11 34
New York City, New York 91,327 529 0.6 96 4 2 33
Newark, New Jersey 16,153 550 3.4 80 2 0.3 21
Norfolk, Virginia 10,259 580 5.7 86 4 14 9
Philadelphia, Pennsylvania 58,722 586 1.0 92 3 25 24
St Louis, Missouri 10,942 633 5.8 83 0.2 n/af 8
San Diego, California 25,946 550 2.1 98 2 39 44
San Francisco, California 28,462 646 2.3 90 6 36 51
San Juan, Puerto Rico 15,031 585 3.9 98 4 2 --
Seattle, Washington 28,505 471 1.7 85 2 6 52
Total/Medianh 726,607 13,519 2.3 90 5 N/A NA

Abbreviations: HIV, human immunodeficiency virus; IDU, injecting drug user; NHBS-IDU, National HIV Behavioral Surveillance System: Injecting Drug Users; n/a, not applicable.
a Number of IDUs in the MSA was obtained from Brady et al. [17].
b "Sample size" includes all recruited persons regardless of eligibility.
c Sampling fraction was calculated as the NHBS-IDU sample size (column 2) divided by the IDU population size (column 1).
d Denominators do not include records without recruitment information, lost records, or persons excluded based on validity of response or gender.
e Percentage of non-seed participants who said the person who gave them the coupon was a stranger.
f Did not use multiple field sites.
g Race/ethnicity cross-recruitment not calculated for San Juan as 99% were Hispanic. The Norfolk sample was 87% Black and the St Louis sample was 91% Black.
h Total for population size, sample size, proportion eligible, and proportion recruited by a stranger; median value for sampling fraction.