Volume 2: No.
4, October 2005
Methods and Baseline Characteristics of Two Group-Randomized Trials
With Multiracial and Multiethnic Working-class Samples
Anne M. Stoddard, ScD, Nancy Krieger, PhD, Elizabeth M.
Barbeau, ScD, Gary G. Bennett, PhD, Martha E. Fay, MPH, Glorian Sorensen, PhD,
MPH, Karen Emmons, PhD
Suggested citation for this article: Stoddard AM, Krieger N, Barbeau
EM, Bennett GG, Fay ME, Sorensen G, et al. Methods and baseline characteristics of two
group-randomized trials with multiracial and multiethnic working-class samples. Prev Chronic Dis [serial online] 2005 Oct [date
cited]. Available from: URL: http://www.cdc.gov/pcd/issues/2005/
Few papers address the methodological challenges in
recruiting participants for studies of cancer prevention interventions
designed for multiracial and multiethnic working-class populations. This paper
reports the results of the sample selection and survey methods for two
group-randomized intervention studies.
The two group-randomized intervention studies, Healthy Directions–Small
Business (HD–SB) and Healthy Directions–Health Centers (HD–HC), included a worksite-based
study in 26 small manufacturing businesses and a study in 10 outpatient health
centers. We used selection and recruitment methods to obtain a
multiracial and multiethnic working-class study sample. In 2000 and 2001, we
assessed baseline measures of sociodemographic characteristics and behavioral
outcomes by self-report. We then
correlation coefficients (ICCs).
Of the 1740 participants in the HD–SB study, 68% were
non-Hispanic whites, and 76% had working-class occupations. In the HD–HC study, 59% of 2219 participants were non-Hispanic whites. Among those
who worked, 51% had working-class occupations. Large percentages of both
samples reported not meeting recommended guidelines for the target behaviors.
For example, 86% of members of both samples consumed fewer than the
recommended five servings of fruits and vegetables per day. The ICCs for the
four target behaviors in HD–SB were between 0.006 and 0.02. In the HD–HC
study, the ICCs ranged from 0.0004 to 0.003.
The two studies were successful in recruiting multiracial and multiethnic
working-class participants. Researchers will find the estimates of the
primary outcomes and their ICCs useful for planning future
Back to top
Increasingly, there have been calls for reducing health disparities based
socioeconomic position and race and ethnicity (1) and for implementing community interventions
that address segments of the population in which risk for chronic disease is
concentrated (2,3). Few papers in the literature, however, address
the methodological challenges in recruiting participants for studies of such
interventions. This paper describes and presents the results of the sample
selection and survey methods for two group-randomized trials of cancer
prevention interventions designed for multiracial and multiethnic
The Harvard Cancer Prevention Program Project, Healthy Directions, was
designed to develop and evaluate cancer prevention interventions for
multiracial and multiethnic working-class populations (4). The project comprised
two intervention studies, Healthy Directions–Small Business (HD–SB) and
Healthy Directions–Health Centers (HD–HC), and a cancer prevention policy model-analysis
project. The intervention projects were group-randomized controlled studies
that tested the shared primary hypotheses that mean levels of dietary and
physical activity outcomes would improve more significantly in the
intervention group than in the control group. The interventions developed for
the two projects were based on a common conceptual framework (4) drawing on
social ecological theory (3,5). Using this framework, the social context in
which people live was incorporated into the design and delivery of the
interventions. This framework encompasses several factors, including individual
factors (e.g., material circumstances), interpersonal factors (e.g., family
roles and responsibilities), organizational factors (e.g., access to health
care), and community factors (e.g., neighborhood safety). In contrast to
interventions designed for a specific racial or ethnic
group, we used this framework to design interventions that were suitable for a
multiracial and multiethnic population.
HD–SB was a worksite-based
intervention study designed to test the effectiveness of an integrated health
promotion and occupational health protection intervention in 26 small
manufacturing businesses in Massachusetts (6). HD–HC was a health-center–based intervention in 10
community health centers in metropolitan Boston (7). The two intervention
studies were aimed at four primary outcomes: increasing fruit and vegetable
consumption, decreasing red meat consumption, increasing daily multivitamin
use, and increasing physical activity. In both studies, the organization was
the unit of randomization and intervention, and the individual worker or
health center member was the unit of observation.
Group-randomized trials are those in which groups of individuals are
randomized to study conditions, but observations are made on the individuals
within the groups (8). An advantage of this design is the ability to enhance
the intervention effectiveness through the social interactions among members
of the groups randomized. The main disadvantage is the loss of statistical
efficiency due to the correlation in behavior among members of the same group
(9). This study design has been increasing in popularity over the last 25
years, especially for the evaluation of community-based interventions (10).
Planning for such studies requires estimates of the within-group correlation of the proposed
outcome measures, yet published estimates for specific behaviors and
populations are hard to find because there are few publications that include
these values in reports of results.
This report focuses on our success in recruiting multiracial and
multiethnic working-class participants. We compare the characteristics of the participants
with selected characteristics of the larger population within which they reside, and we provide
point estimates of the outcome measures and estimates of the intraclass
correlation coefficients (ICCs). The ICC is the fraction of the total
variation in a measure that is attributable to the clustering of the behavior
by members of the same group in comparison with members of different groups
(i.e., the health center or worksite).
This information is important for researchers planning group-randomized trials
in diverse working-class populations.
Back to top
The methods of both studies were approved by Dana-Farber Cancer Institute’s
Office for the Protection of Research Subjects and the Harvard School of
Public Health’s Human Subjects Committee. Additionally, the methods of the
small business study were approved by Beth Israel Deaconess Medical Center’s
Committee on Clinical Investigations, and the methods of the health centers
study were approved by Harvard Vanguard Medical Associates–Department
of Ambulatory Care and Prevention.
For HD–SB, we identified 224 worksites through D&B
(The D&B Corp, Short Hills, NJ; www.dnb.com*) listings of manufacturing
businesses with Standard Industrial Classification (SIC) codes 20–39 (U.S.
Department of Labor, Occupational Safety & Health Administration,
Washington, DC; www.osha.gov/pls/imis/sicsearch.html) located in the
metropolitan Boston area and employing between 30 and 150 workers. Businesses
with these SIC codes were selected because they are more likely than those in
other sectors to use potential carcinogens in work processes and thereby are
suitable for cancer prevention interventions that integrate health protection
and health promotion.
Further eligibility criteria included: 1) employing a
multiracial and multiethnic population, defined as 25% of workers being first- or
second-generation immigrants or people of color; 2) having an employee turnover rate of
less than 20% in the previous year; and 3) being autonomous in decision-making
power to participate in the study if part of a larger parent company. Of the
224 businesses initially identified, 197 (88%) completed the prerecruitment
survey assessing these eligibility criteria and, of these, 131 (66%) met
Finally, companies had to consent to being randomized to receive the
behavioral and occupational health intervention and to provide time at work
for employees to complete assessment surveys and to participate in the
intervention activities. Of the 131 eligible companies, 26 (20%) consented to
participate in the study. Details of the recruitment process and comparison of
worksites recruited and not recruited are provided elsewhere (11).
Worksites ranged in size from 32 to 137 workers. All employees who met the
following criteria were eligible to receive the interviewer-administered
survey: 1) permanent employee, 2) worked 20 hours or more per week, 3) worked
onsite, and 4) spoke English, Spanish, Portuguese, or Vietnamese. Interviews
were conducted in English, Spanish, Portuguese, and Vietnamese between May and
December 2000. Of 2096 eligible employees, 1740 (83%) completed the survey.
Harvard Vanguard Medical Associates, a 14-center
multispecialty medical group practice serving more than 270,000 patients in
the greater Boston area, provided the venues for the HD–HC study. We
selected the 10 health centers with the most racial, ethnic, and socioeconomic
diversity for this study. A random sample of health center members was
selected from each center using a list of eligible patients and a random
number generator. Eligibility criteria included: 1) living in an
eligible neighborhood (see below); 2) being 18 to 75 years old; 3) having a
well-care or follow-up visit scheduled with a participating provider; 4) being
able to speak and read either English or Spanish (unlike the worksites,
Portugese and Vietnamese were not commonly spoken languages); 5) not having cancer at the
time of enrollment; and 6) not being employed by the participating health
centers or a worksite participating in the small business study. Eligible
neighborhoods were defined as census block groups that were predominantly
working class (66% or more of employed persons are in working-class
occupational groups comprised predominantly of nonsupervisory employees); or
met the federal definition of a “poverty area” (20% or more of the
population lives below the poverty line); or had low levels of education (25%
or more of the adult population has not completed high school) (12).
All 117 providers (physicians, nurse practitioners, and physician
assistants) practicing in the internal medicine departments of those centers
were approached for permission to recruit from among their patients. A total
of 97 (83%) of the 177 clinicians participated, with no differences in the rates of
clinician participation between the intervention and control conditions.
We identified patients in the eligible age range who were
scheduled for appointments with one of the participating providers through the
health center’s automated central appointment system. To determine whether a
potential participant lived in an eligible neighborhood, the residential
address was geocoded to the census block group, a subdivision of the census
tract and the smallest census geographic area (approximately 1000 people) that
provides socioeconomic data. Socioeconomic data from the 1990 Census were used
to identify eligible neighborhoods. Geocoding was conducted by a commercial
firm with verified high accuracy (96%) (13).
Potential participants received a letter describing the study and providing
a number to call if they did not want to participate. Members who did not
reply within 2 weeks were then contacted by telephone, and after their
eligibility was confirmed, they were invited to participate. If they
consented, they completed the oral survey at that time or made an appointment to be
interviewed by telephone at another time. Study staff attempted to recruit
8963 potentially eligible candidates during 2000 and 2001. Of these, 2547 were
unreachable. Among the 6416 who were reached, 867 (14%) were ineligible; 3330
(52%) refused to participate and 2219 (35%) were enrolled. Assuming that 14% of those
unreachable were also ineligible, the response rate is 29% of those assumed
Each survey included a core set of items in addition
to items unique to that project which reflected mediating and moderating
We assessed three dimensions of
socioeconomic position (education, poverty status, and occupational class) and
two dimensions of race and ethnicity (racial or ethnic identification and
whether the respondent and his or her parents were born in the United States).
Respondents reported their educational level in nine categories, which we
subsequently collapsed to four (did not complete high school, high school
diploma or equivalent, some post-high-school training, and baccalaureate
degree or more). Household income was assessed in $10,000 increments from less
than $10,000 per year to $50,000 per year or more. We combined the responses to this
item with number of people supported by the income and the ages of household
members to categorize respondents according to the federal poverty guidelines
for food aid (14). In 2001, the poverty guideline for a single person was
$9,214; for a family of two adults and two children it was $17,960. The
guideline for eligibility for food stamps and The Special Supplemental
Nutrition Program for Women, Infants, and Children (WIC) is no more than 185% of the
poverty guideline. Respondents were classified as below the poverty guideline,
above the poverty guideline but below 185% of the guideline, or above 185% of
the poverty guideline.
We combined information about the respondent’s current or most recent job
title into a three-category occupational class variable: working class
(clerical, sales, skilled or unskilled labor), professional/managerial
(professional, managerial, or technical), or no job title. This latter group
included health center participants who were homemakers, disabled, and others
who were not in the paid labor force and did not report a recent job title.
Participants were asked whether they were of Hispanic or Latino heritage
and whether they belonged to any of the four racial groups. We coded participants
who reported being of Hispanic or Latino origin in the Hispanic group
regardless of any other responses. For the rest, those who reported only one
racial group were categorized in that group (i.e., American Indian or Alaska
Native, Asian or Pacific Islander, black or African American, or white).
Respondents who selected more than one racial group were classified as
multiple heritage and were subsequently classified as those who included white
and those who did not.
We combined information about the participants’ and their parents’ birth
places into the following three-category measure of immigration status:
participant born outside the United States (defined as outside the 50 states and the
District of Columbia), participant born in the United States but one or more parents
born outside the United States, and participant and both parents born in the
Respondents were also asked their birth date and sex.
The target levels of the health behaviors, based
on well-established recommendations (1,15,16), were: five or more servings of
fruit and vegetables per day, three or fewer servings of red meat per week,
daily multivitamin use, and at least 2.5 hours of moderate or vigorous
physical activity per week. For each
of the target behaviors we dichotomized the continuously scaled summary
measures at the intervention target level so that we could compute the
percentage of participants who met the intervention target.
Servings of fruits and vegetables consumed per day were assessed using a screener (17-19)
that asked about usual consumption over
the last 4 weeks of seven common foods (orange and grapefruit juice, other
fruit juice, green salad, fried potatoes, potatoes other than fried, fruit,
and other vegetables). For each food, respondents chose 1 of 10 precoded
responses from never to five or more times per day. The responses were recoded
to equivalent servings per day and summed to obtain total fruit and vegetable
servings per day. We then computed a dichotomous measure of either five or
more servings per day or less than five servings per day.
Servings of red meat were assessed using an abbreviated form of the
semiquantitative food frequency questionnaire (20). The screener asked about
usual consumption over the last 4 weeks of six common foods (processed
meat; hamburger; beef, ham, pork, or lamb in a sandwich or mixed dish; 4 to 6
ounces of beef, ham, pork, or lamb as a main dish; 4 to 6 ounces of poultry;
and 3 to 5 ounces of fish). The six response categories ranged from never to
one or more times per day. The responses were recoded to equivalent servings
per week and summed for total servings of red meat per week. The totals were
dichotomized to three or fewer servings or more than three servings per week.
We based our physical activity assessment on the questionnaire used in the
Nurses’ Health Study (21). We asked how often on average in the last four
weeks respondents engaged in each of eight moderate or vigorous leisure
activities. We adapted the items to include specific activities that might be
more common in the study population. Activities included walking for exercise;
jogging; running; bicycling; aerobics or aerobic dancing; lifting weights;
playing soccer, rugby, basketball, lacrosse, baseball, or football; or other
activities that get the respondent out of breath. There were eight response
categories ranging from never to more than 6 hours per week. In addition, we
asked about usual walking pace. The responses were recoded to equivalent
minutes per week and summed for total minutes of physical activity per week.
Walking was included if usual pace was reported to be faster than “easy,
casual.” The sum was collapsed to 150 minutes (2.5 hours) or more per week or more
compared with fewer than 150 minutes per week.
Furthermore, we asked respondents on average how many days they take a
multivitamin. Respondents were coded as taking a multivitamin daily if they
reported taking one 6 or 7 days per week.
For each study sample, we report the number and percentage of participants
according to the measures of sociodemographic characteristics and their levels
of health behaviors. For comparison purposes, we also present available 2000
census data for the consolidated
metropolitan statistical area (CMSA) covering eastern Massachusetts (22). We
present the sex distribution for the population aged 18 years and older,
educational attainment for the population aged 25 years and older,
occupational class for the employed population aged 16 years and older, and
percentage below the poverty line for individuals aged 18 years and older. We
report race and ethnicity (Hispanic and non-Hispanic white) and percentage
of non-U.S.–born for the population as a whole.
For each health behavior, we computed the adjusted percentage of
respondents who practice the behavior, controlling for the clustering of
participants in randomization units, health centers, or worksites. We also
computed the ICC of each health behavior in each study. The adjusted
percentages and ICCs were computed using linear logistic regression analysis
with group (health center or worksite) as a random effect (8).
Computations were carried out using the GLIMMIX macro to the SAS
statistical software (SAS Institute Inc, Cary, NC) (23,24).
Back to top
Sociodemographic characteristics of the two samples
Table 1 shows the sociodemographic
characteristics of the two samples and of the greater Boston area. The
population of the eastern Massachusetts CMSA is 81%
non-Hispanic white, compared with 68% of the HD–SB sample and 59% of the HD–HC
sample. The HD–SB sample included 13% Hispanic or Latino ethnicity and
approximately equal percentages of blacks (5%) and Asians (7%). In the HD–HC
sample, 26% were black and 8% were Hispanic. About one third (34%) of the HD–SB
participants and 22% of the HD–HC participants were born outside the United
States. Additionally, 10% of the U.S.-born HD–SB participants and 18% of the
participants had a parent or parents who were born outside the United States.
In the eastern Massachusetts CMSA, 41% of the adults have a high school education or less,
slightly less than those in the HD–SB sample (46%) but more than those in the
HD–HC sample (28%). Only 24% of the HD–SB participants were professional,
managerial, or technical workers. The remaining 76% were employed in
working-class occupations (i.e., clerical, sales, skilled or unskilled labor). Among
HD–HC participants, approximately equal percentages were employed in
professional, managerial, or technical positions (45%) and in working-class
occupations (44%). In the greater Boston area, 57% of employed adults are
employed in working-class occupations. Although most of the participants in
both studies were at or above 185% of the poverty guideline, 15% of HD–SB
participants were below this cut point, even though they were all employed. In
the HD–HC sample, 18% of participants were below 185% of poverty.
At baseline, most participants in both studies did not meet the
intervention targets for fruit and vegetable consumption and daily
multivitamin use (Table 2). In the HB–SB study, most participants did not
meet the target for red meat consumption, but among HB–HC participants,
almost half met that target. Surprisingly high percentages (73% for HB–SB
and 65% for HB–HC) of participants reported at least 2.5 hours of physical
activity per week in both studies.
Estimates of ICCs for the primary outcomes
Table 3 presents the adjusted prevalence of each health behavior
controlling for the clustering of respondents in randomization units, along
with the ICC. The adjusted prevalences of the target behaviors are very
close to the unadjusted prevalences presented in Table 2. The ICCs for the
four target behaviors in HD–SB were between 0.006 and 0.02, indicating a
small level of concordance among workers in the same worksites. In the HD–HC
study, the ICCs were considerably smaller, ranging from 0.0004 to 0.003.
Back to top
These two studies were successful in sampling a multiracial and multiethnic
subpopulation of eastern Massachusetts residents. The sampling strategies of
both studies reached a subpopulation that is more heterogeneous in racial and
ethnic make-up than the greater Boston area. Furthermore, the HD–SB sample
has a larger percentage of members with working-class occupations and those
with a high school education or less than the general population of adults.
disparities in the United States are often described in terms of racial or ethnic
inequalities; yet, within racial and ethnic groups, there is
variability in both socioeconomic position and morbidity and mortality risk.
Nevertheless, populations of color bear a disproportionate burden of poverty
(25-27). It is well known that socioeconomic deprivation adversely affects
health and increases mortality (12,28). The concepts of social class and
socioeconomic position are complex and encompass occupational class, income,
poverty, wealth, education, and prestige or status at the individual,
household, and area levels (12). We have measured two dimensions of race and
ethnicity and three of socioeconomic position. Maintaining these separate
characteristics, rather than attempting to define a single measure of
socioeconomic position, will allow us to explore the interactions among them
in understanding the determinants of successful interventions.
Small percentages of the study sample respondents lived in households that
were below the poverty threshold, as expected among a population of working-class participants and
those with health insurance. Nevertheless, a substantial proportion of our
samples would be eligible for food aid — 15% in the HD–SB sample and 18% in
the HD–HC sample. These categorizations do not take into consideration
regional differences in cost of living. The greater Boston area is one of the
most expensive areas in the country; the self-sufficiency standard for a
family of four in that area was $42,564 in 1998 and $54,612 in 2003 (29).
Although both study samples represent multiracial and multiethnic
working-class populations, the two samples differ from one another. The HD–SB sample was
somewhat younger and included more men than the HD–HC sample. The HD–SB
sample had a higher percentage of Asian and Hispanic respondents and a higher
percentage of recent immigrants than the HD–HC sample. The HD–HC sample
had a higher percentage of participants with household incomes below 185% of
poverty than the HD–SB sample. The HD–SB sample has a higher percentage of
respondents with less than a high school education than the HD–HC sample.
The differences between the two study samples in the levels of the health
behaviors reported by the participants reflect these sociodemographic
differences. The workers in the HD–SB sample were more physically active
than those in the HD–HC sample were, and a higher percentage of the HD–HC
sample members reported taking a multivitamin daily. These differences may be
attributable to the differences in age, sex, education, or other factors.
Despite these differences in health behavior practices, the percentages of
respondents in both samples who were at lower levels of the target behaviors
are high, indicating a need for the behavior change interventions in both
The mean hours of physical activity reported by members of both samples is
surprisingly high. Although we asked about leisure time activity, participants
may have conflated their reports to include activities related to occupational
activities, domestic chores, childcare, and walking for transportation, for
example. A small validity study done in conjunction with this project
indicated, however, that total hours of activity were reported accurately
(data not shown).
The response rate to the baseline survey in HD–HC was low, due in part to
the fact that potential participants were agreeing to participate in a
randomized trial, not just a health survey. This is similar to the low
recruitment rate for the businesses in the HD–SB study. The internal
validity of the intervention trials is assured by randomization, and the
survey is a valid baseline assessment of the levels of behaviors in the two
intervention groups prior to intervention implementation. For other
researchers who might use the baseline measures and ICCs to plan studies, the
generalizability to health center members and workers in worksites who would
consent to participate in such a study is also appropriate.
Planning for group-randomized trials of the effectiveness of interventions
targeting modifiable health behaviors requires estimates of the means,
variances, and ICCs of the behaviors within the study population of interest
(8,10). The estimates reported here apply to four specific health behaviors
and two types of randomization groups. The estimated ICCs in the HD–SB
sample are similar to those found in other worksite-based intervention studies
(30). The estimated ICCs in the HD–HC sample are lower than those in the HD–SB
sample are but are similar to those at the district health authority level in
England (31). Nevertheless, the ICCs in both studies are sufficient to
influence the error variance of the test statistic for evaluating the
effectiveness of the intervention and must be included in power calculations
for group-randomized studies.
In summary, the procedures developed by these two intervention studies to
sample multiracial and multiethnic working-class populations in eastern
Massachusetts were successful in identifying such groups. These samples are
more diverse in their racial and ethnic make-up and other sociodemographic
characteristics than the greater Boston population. Although the
subpopulations resided in the same geographic area and may overlap in other
ways, the HD–HC sample was restricted explicitly to exclude anyone in the HD–SB
sample. Despite the close proximity of these two subpopulations, they differ
in ways that would be expected by their provenance. Furthermore, both samples
represent populations with high percentages of members who have cancer-related
There has been a call for research on the effectiveness of interventions
targeting modifiable health behaviors (28), yet intervention approaches have
not been designed for or sufficiently tested in working class, ethnically
diverse populations (32). Our explicit aim was to recruit from the large
multiracial and multiethnic group of working-class men and women at elevated risk
for adverse health outcomes. This group is confronted with constraints and
limited resources that may influence patterns of health behaviors. We have
developed behavioral interventions that respond to the social contextual
realities of this group as an approach to reducing the excess burden of
cancer borne by communities of color and lower socioeconomic position (4). To
fully evaluate the effectiveness of these interventions, it is important to
study a diverse population. Future manuscripts will report on the
effectiveness of the interventions in promoting change in the behaviors and
the influence of the social context on behavior and behavior change.
Back to top
This work was supported by a grant from the National Cancer
Institute (P01 CA75308). The studies and analysis reported were the product of
the authors and not the NCI.
Back to top
Corresponding Author: Anne M. Stoddard, ScD, New England Research
Institutes, 9 Galen St, Watertown, MA 02472. Telephone: 617-932-7747, ext 331.
Author Affiliations: Nancy Krieger, PhD, Martha E. Fay, MPH, Department of Society,
Human Development and Health, Harvard School of Public Health, Boston, Mass; Elizabeth M. Barbeau, ScD, Gary G. Bennett, PhD, Glorian Sorensen, PhD, MPH, Karen Emmons,
PhD, Department of Society, Human Development and Health, Harvard School of Public
Health, and Center for Community Based Research, Dana-Farber Cancer Institute,
Back to top
- U.S. Department of Health and Human Services. Healthy People 2010:
understanding and improving health. 2nd ed. Washington (DC): U.S. Government
Printing Office; 2000 Nov.
- Prothrow-Stith D, Gibbs B, Allen A.
Reducing health disparities: from
theory to practice. Cancer Epidemiol Biomarkers Prev 2003;12(3):256-60.
- Stokols D.
Translating social ecological theory into guidelines for
community health promotion. Am J Health Promot 1996;10(4):282-98.
- Sorensen G, Emmons K, Hunt MK, Barbeau E, Gorman R, Peterson K, et al.
A model for incorporating the social context in health behavior interventions:
application for cancer prevention for working class multi-ethnic populations. Prev Med 2003;37:188-97.
- McLeroy K, Bibeau D, Steckler A, Glanz K.
An ecological perspective on
health promotion programs. Health Educ Q 1988;15(4):351-77.
- Hunt MK, Stoddard AM, Barbeau E, Goldman R, Wallace L, Gutheil C, et
Cancer prevention for working class multi-ethnic populations through small
businesses: the Healthy Directions Study. Cancer Causes Control
- Emmons K, Stoddard AM, Gutheil C, Suarez EG, Lobb R, Fletcher R.
prevention for working class multi-ethnic populations through health centers:
the Healthy Directions Study. Cancer Causes Control 2003;14(8):727-37.
- Murray DM. Design and
analysis of group randomized trials. New York (NY):
Oxford University Press; 1998.
- Donner A, Klar N.
Pitfalls of and controversies in cluster randomization
trials. Am J Public Health 2004;94(3):416-22.
- Donner A, Klar N. Design and analysis of cluster randomization trials in
health research. London: Arnold; 2000.
- Barbeau EM, Wallace L, Lederman R, Lightman N, Stoddard A, Sorensen G.
Recruiting small manufacturing worksites that employ multiethnic, low-wage
workforces into a cancer prevention research trial. Prev Chronic Dis
[serial online] 2004 Jul [cited 2005 Jun 29].
- Krieger N, Williams DR, Moss NE.
Measuring social class in US public
health research: concepts, methodologies, and guidelines. Annu Rev
Public Health 1997;18:341-78.
- Krieger N, Waterman P, Lemieux K, Zierler S, Hogan JW.
On the wrong side of the tracts? Evaluating the
accuracy of geocoding in public health research. Am J Public
- U.S. Department of Health and Human Services. The 2001 HHS poverty guidelines.
Washington (DC): U.S. Government Printing Office; 2003.
- U.S. Department of Health and Human Services. Nutrition and your health:
dietary guidelines for
Americans. 5th ed. Washington (DC): U.S. Government Printing Office; 2000.
- World Health Organization. Diet, nutrition and prevention of chronic diseases:
report of the
joint WHO/FAO expert consultation. Geneva, Switzerland: World Health
- Serdula MK, Coates R, Byers T, Mokdad A, Jewell S, Chávez N, et al.
Evaluation of a
brief telephone questionnaire to estimate fruit and vegetable
consumption in diverse study populations. Epidemiology 1993;4(5):455-63.
- Serdula MK, Coates RJ, Byers T, Simoes E, Mokdad AH, Subar AF.
and vegetable intake among adults in 16 states: results of a brief
telephone survey. Am J Public Health 1995;85(2):236-9.
- Subar AF, Heimendinger J, Patterson BH, Krebs-Smith SM, Pivonka E,
vegetable intake in the United States: the baseline
survey of the 5 A Day for Better Health Program. Am J Health
- Schatzkin A, Freedman LS, Lanza E, Tangrea J.
Diet and colorectal
cancer: still an open question. J Natl Cancer Inst 1995;87(23):1733-5.
- Wolf AM, Hunter DJ, Colditz GA, Manson JE, Stampfer MJ, Corsano KA, et
validity of a self-administered physical activity
questionnaire. Int J Epidemiol 1994;23(5):991-9.
- U.S. Census Bureau. Census 2000 summary file. Washington (DC): U.S. Census Bureau; 2003.
- Littell RC, Milliken GA, Stroup WW, Wolfinger RD. SAS system for mixed
models. Cary (NC): SAS Institute, Inc; 1996.
- Wolfinger R. Laplace’s approximation for nonlinear mixed models. Biometrika 1993;80(4):791-5.
- Fried VM, Prager K, MacKay AP, Xia H. Chartbook on trends in health of
Americans. Health, United States, 2003. Hyattsville (MD): National Center for Health Statistics; 2003.
- Kington RS, Nickens HW. Racial and ethnic differences in health: recent
trends, current patterns, future directions. In: Smelser NJ, Wilson WJ,
editors. America becoming: radial trends and their consequences. Washington (DC): National Academy Press; 2001. p. 253-310.
- Pamuk E, Makuc D, Heck K, Reuben C, Lochner K. Socioeconomic status and
health chartbook. Health, United States, 1998. Hyattsville (MD): National Center for
Health Statistics; 1998.
- Ibrahim SA, Thomas SB, Fine MJ.
Achieving health equity:
journey. Am J Public Health 2003;93(10):1619-21.
- Pearce D, Brooks J. The self-sufficiency standard for Massachusetts.
Boston (MA): Women’s Educational and Industrial Union; 2003 Apr.
- Feng Z, Diehr P, Yasui Y, Evans B, Beresford S, Koepsell TD.
community-level variance in group randomized trials. Stat Med
- Gulliford MC, Ukoumunne OC, Chinn S.
Components of variance and intraclass correlations for the design of community-based surveys and
intervention studies: data from the Health Survey for England 1994. Am J Epidemiol 1999;149(9):876-83.
- Institute of Medicine Committee on Communication for Behavior Change in
the 21st Century. Improving health of diverse populations, speaking of health:
assessing health communication strategies for diverse populations. Washington
(DC): National Academy Press; 2002.
Back to top