ReviewE ffe c ts o f N u rs e -M a n a g e d P ro to c o.docx

Review
E ffe c ts o f N u rs e -M a n a g e d P ro to c o ls in th e O u
tp a tie n t M a n a g e m e n t o f
A dults W ith C h ro n ic C onditions
A System atic Review and M eta-analysis
R yan J. S h a w , P h D , RN; J e n n ife r R. M c D u f f ie ,
PhD ; C ris tin a C. H e n d rix , D N S , NP; A lis o n Edie, D
N P , FNP; L in d a L in d s e y -D a v is , P h D , RN;
A v is h e k N a g i, M S ; A n d rz e j S. K o sin ski, PhD ; an
d Joh n W . W illia m s Jr., M D , M H S c
Background: C h an ges in fe d e ra l h e a lth p o lic y are p ro
v id in g m o re
access t o m ed ica l care f o r persons w ith c h ro n ic
disease. P ro v id in g
q u a lity care m a y re q u ire a te a m a p p ro a c h , w h ic
h th e A m e ric a n
C o lle g e o f Physicians calls th e "m e d ic a l h o m e ." O n
e n e w m o d e l
m a y in v o lv e n u rs e -m a n a g e d p ro to cols.
Purpose: T o d e te rm in e w h e th e r n u rs e -m a n a g e d
p ro to c o ls are e f -
fe c tiv e f o r o u tp a tie n t m a n a g e m e n t o f a d u lts
w ith diabetes, h y p e r-
te n s io n , an d h y p e rlip id e m ia .
Data Sources: MEDLINE, C o c h ra n e C e n tra l R egister o f
C o n tro lle d
Trials, EMBASE, a n d CINAHL fro m Jan ua ry 1 9 8 0 t h ro
u g h January

2 0 1 4 .
Study Selection: T w o review e rs used e lig ib ility c rite ria
t o assess all
title s , ab stracts, a n d fu ll te x ts an d resolved dis a g re e
m e n ts by dis-
cussion o r b y c o n s u ltin g a th ird review e r.
Data Extraction: O n e re v ie w e r d id d a ta a b s tra c tio n
s a n d q u a lity
assessments, w h ic h w e re c o n firm e d b y a s econd
review e r.
Data Synthesis: F rom 2 9 5 4 studies, 1 8 w e re in c lu d e d
. A ll studies
used a reg istere d nurse o r e q u iv a le n t w h o titra te d m
e d ic a tio n s by
f o llo w in g a p ro to c o l. In a m e ta-a na lysis, h e m o g lo
b in A 1c level d e -
creased b y 0 .4 % (9 5 % C l, 0 .1 % t o 0 . 7 % ) (n = 8);
systolic and
d ia s to lic b lo o d pressure decreased b y 3 .6 8 m m H g
(C l, 1 .0 5 to
6.31 m m H g ) an d 1 .5 6 m m H g (C l, 0 .3 6 t o 2 .7 6 m
m H g),
re s p ective ly (n = 12); to ta l cho le s te ro l level decreased b
y 0 .2 4
m m o l/L (9 .3 7 m g /d L ) (C l, 0 . 5 4 - m m o l/L decrease
t o 0 .0 5 - m m o l/L
increase [ 2 0 .7 7 - m g / d L decrease t o 2 . 0 2 - m g / d L
increase]) (n = 9);
a n d lo w -d e n s ity -lip o p ro te in c h o le ste rol level
decreased b y 0.31
m m o l/L (1 2 .0 7 m g /d L ) (C l, 0 . 7 3 - m m o l/L
decrease t o 0 .1 1 - m m o l/L

increase [ 2 8 .2 7 - m g / d L decrease t o 4 . 1 3 - m g / d L
increase]) (n = 6).
Limitation: Studies had lim ite d de s c rip tio n s o f th e in
te rv e n tio n s an d
p ro to c o ls used.
Conclusion: A te a m a p p ro a c h t h a t uses n u rs e -m a n
a g e d p ro to c o ls
m a y ha ve p o s itiv e e ffe c ts o n th e o u tp a tie n t m a
n a g e m e n t o f a d u lts
w ith c h ro n ic c o n d itio n s , such as diabetes, h y p e rte n
s io n , an d
h y p e rlip id e m ia .
Primary Funding Source: U.S. D e p a rtm e n t o f V e te ra n s
A ffa irs.
Ann Intern Med. 2014;161:113-121. d o i:10.7 326 /M 13 -256 7
www.annals.org
For author affiliations, see end o f text.
M edical management of chronic illness consumes 75% of every
health care dollar spent in the United States
(1). Thus, provision of economical and accessible— yet
high-quality— care is a major concern. Diabetes mellitus,
hypertension, and hyperlipidemia are prime examples of
chronic diseases that cause substantial morbidity and mor-
tality (2, 3) and require long-term medical management.
For each of these disorders, most care occurs in outpatient
settings where well-established clinical practice guidelines
are available (4—7). Despite the availability o f these guide-
lines, there are important gaps between the care recom-
mended and the care delivered (8-10). The shortage of
primary care clinicians has been identified as 1 barrier to

the provision of comprehensive care for chronic disease
(11, 12) and is an impetus to develop strategies for expand-
ing the roles and responsibilities o f other interdisciplinary
team members to help meet this increasing need.
The patient-centered medical home concept was de-
veloped in an effort to serve more persons and improve
chronic disease care. It is a model of primary care transfor-
mation that builds on other efforts, such as the chronic
care model (13), and includes the following elements:
patient-centered orientation toward the whole person,
team-based care coordinated across the health care system
and community, enhanced access to care, and a systems-
based approach to quality and safety. Care teams may in-
clude nurses, primary care providers, pharmacists, and be-
w w w .annals.org
havioral health specialists. An organizing principle for care
teams is to utilize personnel at the highest level of their skill
set, which is particularly relevant given the expected in-
crease in demand for primary care services resulting from
the Patient Protection and Affordable Care Act.
W ith this increased demand, the largest health care
workforce, registered nurses (RNs), may be a valuable asset
alongside other nonphysician clinicians, including physi-
cian assistants, nurse practitioners, and clinical pharma-
cists, to serve more persons and improve chronic disease
care. Robust evidence supports the effectiveness o f nurses
in providing patient education about chronic disease and
secondary prevention strategies (14-19). W ith clearly de-
fined protocols and training, nurses may also be able to
order relevant diagnostic tests, adjust routine medications,
and appropriately refer patients.
O ur purpose was to synthesize the current literature

describing the effects o f nurse-managed protocols, includ-
S ee a ls o :
E d ito r ia l c o m m e n t
.....................................................................153
W e b - O n ly
S u p p le m e n t s
C M E q u iz
15 July 2014 Annals of Internal Medicine I Volume 161 •
Number 2 [ 1 1 3
R e v i e w Nurse-Managed Protocols in Managing Outpatients
W ith Chronic Conditions
Figure 1. S u m m a r y o f e v id e n c e s e a rc h a n d s e
le c tio n .
I n c l u d e d ( n = 2 0 )
U n i q u e s t u d i e s : 1 8
C o m p a n i o n a r t i c l e s : 2 *
* Methods or follow-up articles.
ing medication adjustment, for the outpatient manage-
ment o f adults with common chronic conditions, namely
diabetes, hypertension, and hyperlipidemia.
M e t h o d s

W e followed a standard protocol for all steps o f this
review. A technical report that fully details our methods
and presents results for all original research questions
is available at www.hsrd.research.va.gov/publications/esp
/reports.cfm.
D a t a S o u r c e s a n d S e a r c h e s
In consultation with a master librarian, we searched
M ED LIN E (via PubMed), Cochrane Central Register of
Controlled Trials, EMBASE, and CINAHL from 1 Janu-
ary 1980 through 31 January 2014 for English-language,
peer-reviewed publications evaluating interventions that
compared nurse-managed protocols with usual care in
studies targeting adults with chronic conditions (Supple-
ment 1, available at www.annals.org).
W e selected exemplary articles and used a Medical
Subject Heading analyzer to identify terms for “nurse pro
tocols.” W e added selected free-text terms and validated
search terms for randomized, controlled trials (RCTs) and
quasi-experimental studies, and we searched bibliographies
o f exemplary studies and applicable systematic reviews for
missed publications (15, 17, 2 0 -2 9 ). To assess for publi-
cation bias, we searched ClinicalTrials.gov to identify com-
pleted but unpublished studies meeting our eligibility
criteria.
S t u d y S e l e c t i o n , D a t a E x t r a c t i o n , a n d Q u
a l i t y
A s s e s s m e n t
Two reviewers used prespecified eligibility criteria to
assess all titles and abstracts (Supplement 2, available at
1 1 4 15 July 2014 Annals o f Internal Medicine Volume 161 •

Number 2
www.annals.org). Eligibility criteria included the involve-
ment of an RN or a licensed practical nurse (LPN) func-
tioning beyond the usual scope of practice, such as adjust-
ing medications and conducting interventions based on a
written protocol. Potentially eligible articles were retrieved
for further evaluation. Disagreements on inclusion or ex-
clusion were resolved by discussion or a third reviewer.
Studies excluded at full-text review are listed in Supple-
ment 3 (available at www.annals.org). Abstraction and
quality assessment were done by 1 reviewer and confirmed
by a second. We piloted the abstraction forms, designed
specifically for this review, on a sample of included articles.
Key characteristics abstracted included patient descriptors,
setting, features of the intervention and comparator, match
between the sample and target populations, extent of the
nurse interventionist’s training, outcomes, and quality ele
ments. Supplements 4 and 5 (available at www.annals.org)
summarize quality criteria and ratings, respectively.
Because many studies were done outside the United
States, we queried the authors o f such studies about the
education and scope of practice o f the nurse intervention-
ists. Authors were e-mailed a table detailing the credential-
ing and scope of practice of various U.S. nurses and asked
to classify their nurse interventionist.
D a t a S y n t h e s i s a n d A n a l y s i s
The primary outcomes were the effects of nurse-
managed protocols on biophysical markers (for example,
glycosylated hemoglobin or hemoglobin A lc [HbAlc]), pa-
tient treatment adherence, nurse protocol adherence,
adverse effects, and resource use. W hen quantitative syn-
thesis (that is, meta-analysis) was feasible, dichotomous

outcomes were combined using odds ratios and continuous
outcomes were combined using mean differences in
random-effects models. For studies with unique but con-
ceptually similar outcomes, such as ordering a guideline-
indicated laboratory test, we synthesized outcomes across
conditions if intervention effects were sufficiently homoge-
neous. We used the Knapp and H artung method (30, 31)
to adjust the SEs of the estimated coefficients.
For categories with several potential outcomes (for ex-
ample, biophysical markers) that may vary across chronic
conditions, we selected outcomes for each chronic condi-
tion a priori: H bA lc level for diabetes, blood pressure (BP)
for hypertension, and cholesterol level for hyperlipidemia.
In 1 example (32), we imputed missing SDs using esti-
mates from similar studies.
We computed summary estimates of effect and evalu-
ated statistical heterogeneity using the Cochran Q and I 2
statistics. We did subgroup analyses to examine potential
sources o f heterogeneity, including where the study was
conducted and intervention content. Subgroup analyses in-
volved indirect comparisons and were subject to confound-
ing; thus, results were interpreted cautiously. Publication
bias was assessed using a ClinicalTrials.gov search and fun-
w w w .a n n a ls .o r g
E x c l u d e d a t t h e t i t l e / a b s t r a c t
le v e l ( n = 2 6 1 5 )
E x c l u d e d ( n = 3 1 9 )
N o t E n g l i s h , w e s t e r n i z e d c o u n t r y ,

o r f u l l p u b l i c a t i o n : 5 5
N o a d u l t s w i t h d i s e a s e o f i n t e r e s t
o r c o n d u c t e d in a n o u t p a t i e n t
m e d i c a l s e t t i n g : 2 9
I n e l i g i b l e s t u d y d e s i g n o r
c o m p a r a t o r : 7 5
N o i n t e r v e n t i o n o f in t e r e s t : 1 5 3
N o o u t c o m e o f in t e r e s t : 7
S e a r c h r e s u l t s o f
r e f e r e n c e s ( n = 2 9 5 4 )
R e t r i e v e d f o r
f u l l - t e x t r e v i e w
( n = 3 3 9 )
Nurse-Managed Protocols in Managing Outpatients With
Chronic Conditions R e v i e w
nel plots when at least 10 studies were included in the
analysis.
W hen quantitative synthesis was not feasible, we ana-
lyzed data qualitatively. We gave more weight to evidence

from higher-quality studies with more precise estimates of
effect. The qualitative syntheses identified and documented
patterns in efficacy and safety of the intervention across
conditions and outcome categories. We analyzed potential
reasons for inconsistency in treatment effects across studies
by evaluating variables, such as differences in study popu-
lation, intervention, comparator, and outcome definitions.
W e followed the approach recommended by the
Agency for Healthcare Research and Quality (33) to eval-
uate the overall strength of the body o f evidence. This
approach assesses the following 4 domains: risk o f bias,
consistency, directness, and precision. These domains were
considered qualitatively, and a summary rating o f high,
moderate, low, or insufficient evidence was assigned.
R o le o f th e F u n d in g Source
The Veterans Affairs Quality Enhancement Research
Initiative funded the research but did not participate in the
conduct of the study or the decision to submit the manu-
script for publication.
R e s u l t s
O ur electronic and manual searches identified 2954
unique citations (Figure 1). O f the 23 potentially eligible
studies, 4 were excluded because we could not verify
whether nurses had the authority to initiate or titrate med-
ications and the author did not respond to our query for
clarification (34—37). We excluded a trial of older adults in
which we could not differentiate the target illnesses (38).
Approximately two thirds of the authors we contacted for
missing data or clarification responded.
We included 18 unique studies (23 004 patients) that
focused on patients with elevated cardiovascular risk (Ta-

ble) (32, 3 9 -5 5 ). O f these, 16 were RCTs and 2 were
controlled before-and-after studies on diabetes (49, 53).
The comparator was usual care in all but 1 study, in which
a reverse-control design was used, and each intervention
served as the control for the other. Eleven studies were
done in Western Europe and 7 in the United States. Me-
dian age o f participants was 58.3 years (range, 37.2 to 72.1
years) based on 16 studies. Approximately 47% of the par-
ticipants were female. Race was not reported in 84% o f the
studies. Supplement 5 gives detailed study characteristics.
No outstanding studies were identified through Clinical-
Trials.gov. Supplement 6 provides funnel plots that assess
publication bias (available at www.annals.org).
Overall, these studies displayed moderate risk of bias.
Two studies were judged as having a high risk o f bias
because o f inadequate randomization (44, 53), 12 were
moderate risk (32, 3 9 - 4 1 , 43, 47-52, 54), and 4 were low
risk (42, 45, 46, 55). O ther design issues affecting risk-of-
bias ratings were possible contamination from a concurrent
Table. Study and Patient Characteristics of Included
Diabetes, Hypertension, and Hyperlipidem ia Studies
Characteristic Cardiovascular Risk
Studies, n ( % )
Total
Studies 18
Patients* 23 004
Design
RCT 16 (89)
Non-RCT 2 ( 1 1 )
Location

U nited States 7 ( 3 9 )
W estern Europe 11 (61)
S etting
General medical hospital 12 (67)
Specialty hospital 3 (17)
Primary clinic and specialty hospital 2 ( 1 1 )
Telephone- and clinic-delivered care 1 (5.5)
Inte rv ention
Target
Glucose 15 (83)
Blood pressure 11 (61)
Lipids 9 ( 5 0 )
Delivery
Clinic visits 15 (83)
Primarily telephone 3 ( 1 7 )
D uration
6 m o 2 ( 1 1 )
12 m o 8 (44.5)
> 1 2 m o t 8 (44.5)
Nurse tra in in g
Specialist* 3 ( 1 7 )
Received study-specific tra inin g 10 (55)
Case m anager 1 (5.5)
N o t described 4 ( 2 2 )
M e d ic a tio n in itia tio n 11 (61)
Education or behavioral strategy

Education 1 6 (8 9 )
Specific behavioral s tra te g y ! 3 ( 1 7 )
Self-m anagem ent plan 9 ( 5 0 )
O u tc o m e
H em oglobin A 1c level 12 (67)
Blood pressure 14 (78)
Cholesterol level 1 5 (8 3 )
Performance measure 13 (72)
Behavioral adherence 4 ( 2 2 )
Protocol adherence 1 (6)
Risk o f b ia s /q u a lity
L o w /g o o d 4 ( 2 2 )
M o d e ra te /fa ir 12 (67)
H ig h /p o o r 2 (11)
RCT — randomized, controlled trial.
* Number of patients represents the total mean of 22 839 and 23
170 because in
1 included study (30), hypertension and hyperlipidemia results
were reported on 2
different but overlapping populations due to randomization,
t Range, 14-36 mo.
$ Clinical certification or diabetes nurse educator.
§ Motivational interviewing.
w w w .annals.org
15 July 2014 Annals of Internal Medicine Volume 161 •Number
2 1 1 5

F i g u r e 2 . Effects of nurse-managed protocols on
hemoglobin A1c level.
Study, Year (Reference) Nurse Protocols Total, n Usual Care
Total, n
Mean (SD) Mean
A u b e rte ta l, 1 9 9 8 (4 0 ) 7.10 (1.33) 51 8.20
Bellary et al, 2 0 0 8 (4 2 ) 8.20 (1.74) 868 8.35
H ouw e lin g et al, 2009 (47) -1 .5 0 (1.35) 46 -0 .9 0
H ouw e lin g et al, 2011 (46) -0 .0 9 (1.07) 102 0.03
M acM ahon e t al, 2009 (48) -0 .3 4 (0.97) 94 0.12
O 'H are et al, 2004 (52) -0 .2 3 (1.42) 182 -0 .2 0
Taylor e t al, 2003 (32) -1 .1 4 (1.35) 61 -0 .3 5
W allym ahm ed et al, 2011 (54) 9.30 (1.40) 40 9.70
Summary ( /2 = 69 .8% )
(SD)
W eighted Mean
Difference
(95% Cl), %
-1 .1 0 (-1 .6 2 t o -0 .5 8 )

-0 .1 5 (-0 .3 3 to 0.03)
-0 .6 0 (-1 .1 5 t o -0 .0 5 )
-0 .1 2 (-0 .4 3 to 0.19)
-0 .4 6 (-0 .7 4 t o -0 .1 8 )
-0 .0 3 (-0 .3 4 to 0.28)
-0 .7 9 (-1 .2 4 t o -0 .3 4 )
-0 .4 0 (-0 .9 9 to 0.19)
-0 .4 0 (-0 .7 0 t o -0 .1 0 )
intervention, unblinded outcome assessors, and incomplete
outcomes data.
Characteristics o f the Interventions
All 18 study interventions used a protocol and re-
quired the nurse to titrate medications; however, only 11
reported that the nurse was independently allowed to ini-
tiate new medications. All but 1 study (55) provided the
actual algorithm or citation. An RN (not an advanced
practice RN) was the interventionist in all U.S. studies; a
nurse with an equal scope o f practice was the intervention-
ist in the non-U.S. studies. N o studies reported use of
LPNs. In 14 studies, interventions were delivered in a
nurse-led clinic (3 9 -4 2 , 44, 4 6 -5 4 ). Supervisors were
nearly always physicians. O f the studies reporting nurses’
training, 3 used specialists (for example, diabetes-certified),
10 used RNs with study-specific training, and 1 used nurse
case managers with experience in coordinating long-term
care.

Nurse protocols included additional components, such
as education or self-management, in 16 studies. Two stud-
ies (41, 47) did not report additional intervention. Baseline
characteristics showed that patients with diabetes had an
elevated H bAlc level of approximately 8.0% or greater.
Most patients with hypertension had moderate hyperten-
sion, and patients with hyperlipidemia had borderline high
lipid levels. Outcomes were assessed at 6 to 36 months,
with most studies reporting outcomes at 12 months or
longer.
D iabetes O utcom es
O f the 15 studies done in patients with diabetes, 10
RCTs (2633 patients) targeted glucose control. Figure 2
shows the forest plot o f the random-effects meta-analysis
on H bA lc level. Compared with usual care, nurse-managed
protocols decreased H bA lc levels by 0.4% (95% C l, 0.1%
to 0.7%) (n = 8) and effects varied substantially (Q =
23.19; I 2 = 70%). In the 2 non-RCTs (49, 53) not in-
cluded in Figure 2, effects of the protocols on H bA lc level
1 1 6 15 July 2014 Annals o f Internal Medicine Volume 161 •
Number 2
were larger and in the same direction but had higher vari-
ability. Thus, nurse-managed protocols were associated
with a highly variable mean decrease in H bA lc level.
O ther diabetes-related performance measures were
rarely reported (Supplement 6). In 1 controlled before-
and-after study (53), achieving target eye examination, uri-
nary m icroalbumin-creatinine ratio, and foot examination
goals was reported to reach 80% to 100% using nurse-
managed protocols. A second study (49) found a nonsig-
nificant increase in intervention patients achieving eye and
foot examination goals compared with control participants.

Reduction in the proportion of patients with an H bA lc
level o f 8.5% or greater was achieved in 1 study (odds
ratio, 1.69 [Cl, 1.25 to 2.29]) (49).
BP O utcom es
Fourteen studies reported BP outcomes: 13 RCTs
(10 362 patients) and 1 non-RCT (885 patients). Re-
stricted to the 12 RCTs specifically addressing BP (10 224
patients), the intervention decreased systolic BP by 3.68
mm Hg (Cl, 1.05 to —6.31 mm Hg) and diastolic BP by
1.56 mm H g (Cl, 0.36 to 2.76 mm Hg), with high vari-
ability (72 > 70%) (Figures 3 and 4). Funnel plots sug-
gested possible publication bias with systolic but not dia-
stolic BP (Supplement 6). Overall, nurse-managed
protocols were associated with a mean decrease in systolic
and diastolic BP.
Eleven of the 18 studies focused on achieving various
target BPs: 10 RCTs (9707 patients) and 1 non-RCT (885
patients). W hen the analysis was restricted to RCTs, nurse-
managed protocols were more likely to achieve target BP
than control protocols (odds ratio, 1.41 [Cl, 0.98 to
2.02]), but these results could have been due to chance,
and treatment effects were highly variable (Q = 35.20;
/ 2 = 74%) (Supplement 7, available at www.annals.org).
Using the summary odds ratio and median event rate from
the control group of the trials that implemented nurse pro-
tocols, we estimated the absolute treatment effect as a risk
w w w . a n n a l s . o r g
Nurse-Managed Protocols in Managing Outpatients W ith
Chronic Conditions R e v i e w

difference o f 120 more patients achieving target total BP
per 1000 patients (Cl, 6 fewer to 244 more). Funnel plots
suggested some asymmetry but no clear publication bias.
H y p e r l i p i d e m i a O u t c o m e s
Fifteen studies reported hyperlipidemia outcomes: 13
RCTs (14 817 patients) and 2 non-RCTs (1114 patients).
O f these, 9 RCTs (3494 patients) specifically addressed
total cholesterol levels and 6 RCTs specifically addressed
low-density lipoprotein levels (1095 patients). In analyses
restricted to these trials, the intervention was associated
with a decrease in total cholesterol level. Total cholesterol
levels decreased by 0.24 mmol/L (9.37 mg/dL) (Cl, 0.54-
mmol/L decrease to 0.05-mmol/L increase [20.77-mg/dL
decrease to 2.02-mg/dL increase]) [n = 9), and low-
density lipoprotein cholesterol levels decreased by 0.31
mmol/L (12.07 mg/dL) (Cl, 0.73-mmol/L decrease to
0.11-mmol/L increase [28.27-mg/dL decrease to 4.13-
mg/dL increase]) (n = 6), with marked variability in inter-
vention effects (72 > 89%) (Figure 4). Effects o f nurse-
managed protocols on total and low-density lipoprotein
cholesterol levels from the 2 non-RCTs (49, 53) were in
the same direction. Reductions in total cholesterol level
were not statistically significant. Overall, nurse-managed
protocols were associated with a mean decrease in total and
low-density lipoprotein cholesterol levels.
All 11 studies (9221 patients) targeting various total
cholesterol levels were included in the quantitative analysis
(Supplement 7). Nurse-managed protocols were statisti-
cally significantly more likely to achieve target total choles-
terol levels than control protocols (odds ratio, 1.54 [Cl,
Figure 3 . Effects o f n u rs e -m a n a g e d p ro tocols on

systolic (to p ) an d d ia s to lic ( b o tto m ) b lo o d pressure.
Total, n
Mean (SD) Mean (SD)
Bebb et al, 2007 (41) 143.30 (19.50) 743 143.10 (17.70) 677
Bellary et al, 2008 (42) 134.30 (20.36) 868 134.60 (20.36) 618
Denver et al, 2003 (44) 141.10 (19.30) 59 151.00 (21.90) 56
Houweling et al, 2009 (47) -8.60 (20.54) 46 -4.00 (14.91) 38
Houweling et al, 2011 (46) -7.40 (17.82) 102 -5.60 (16.45) 104
MacMahon et al, 2009 (48) -10.50 (17.45) 94 1.70 (19.39) 94
N ew et al, 2003 (51) 147.00 (20.23) 506 149.00 (20.23) 508
New et al, 2004 (50) 142.00 (24.00) 2474 142.17 (24.00) 2531
O'Hare et al, 2004 (52) -6.69 (21.24) 182 -2.11 (17.47) 179
Rudd et al, 2004 (55) -14.20 (16.23) 69 -5.70 (18.59) 68
Taylor et al, 2003 (32) 4.40 (17.45) 61 8.60 (19.39) 66
Wallymahmed et al, 2011 (54) 115.00 (13.00) 40 124.00 (14.00)
41
Summary (/2 = 75.1%)
- 2 0
I “1
-1 5 -1 0 - 5 0
Weighted Mean Difference, mm Hg
Weighted Mean
Difference
(95% Cl), mm Hg
0.20 (-1.73 to 2.13)

-0.30 (-2.40 to 1.80)
-9.90 (-17.46 t o -2.34)
-4.60 (-12.20 to 3.00)
-1.80 (-6.49 to 2.89)
-12.20 (-17.47 t o -6.93)
-2.00 (-4.49 to 0.49)
-0.17 (-1.50 to 1.16)
-4.58 (-8.59 to -0,57)
-8.50 (-14.35 t o -2.65)
-4.20 (-10.61 to 2.21)
-9.00 (-14.88 t o -3.12)
-3.68 (-6.31 t o -1.05)
Total, n
Mean (SD) Mean (SD)
Bebb et al, 2007 (41) 78.20 (10.20) 743 77.90 (10.40) 677
Bellary et al, 2008 (42) 78.40 (8.63) 868 80.31 (8.63) 618
Denver et al, 2003 (44) 79.90 (10.60) 59 82.20 (12.40) 56
Houweling et al, 2009 (47) -1.40 (9.09) 46 -2.40 (7.61) 38
Houweling et al, 2011 (46) -3.20 (10.18) 102 -1.00 (9.26) 104
MacMahon et al, 2009 (48) -5.90 (8.72) 94 -0.51 (9.69) 94
New et al, 2003 (51) 74.00 (11.29) 506 74.79 (11.29) 508
New et al, 2004 (50) 78.20 (16.06) 2474 78.11 (16.06) 2531
O'Hare et al, 2004 (52) -3.14 (10.56) 182 0.28 (10.00) 179
Rudd et al, 2004 (55) -6.50 (10.00) 69 -3.40 (7.90) 68
Taylor et al, 2003 (32) 2.20 (10.00) 61 1.90 (9.30) 66
Wallymahmed et al, 2011 (54) 65.00 (9.00) 40 69.00 (9.00) 41
Summary (/2 = 75.1 %)
Weighted Mean
Difference

(95% Cl), mm Hg
0.30 (-0.77 to 1.37)
-1.91 (-2.80 t o -1.02)
-2.30 (-6.53 to 1.93)
1.00 (-2.57 to 4.57)
-2.20 (-4.86 to 0.46)
-5.39 (-8.03 to -2.75)
-0.79 (-2.18 to 0.60)
0.09 (-0.80 to 0.98)
-3.42 (-5.54 t o -1.30)
-3.10 (-6.12 t o -0.08)
0.30 (-3.07 to 3.67)
-4.00 (-7.92 to -0.08)
-1.56 (-2.76 t o -0.36)
I---------------- 1-----------------
-1 0 - 5 0
Weighted Mean Difference, mm Hg
w w w .a n n a ls .o r g 15 July 2014 Annals o f Internal
Medicine Volume 161 • Number 2 1 1 7
F ig u re 4. E ffe c ts o f n u r s e - m a n a g e d p ro to c o ls
o n t o t a l c h o le s te r o l ( t o p ) a n d l o w - d e n s i t y
lip o p r o t e in c h o le s te r o l ( b o t t o m ) le v e ls .
Study, Year (Reference) Nurse Protocols
Mean (SD)

Total, n Usual Care
Mean (SD)
Total, n
Allison etal, 1999 (39) -19.00 (35.00) 80 -16.00 (35.00) 72
Bellary et al, 2008 (42) 181.50 (26.08) 868 180.35 (26.08) 618
DeBusk etal, 1994 (43) 184.55 (32.05) 243 208.88 (40.54) 244
Houweling et al, 2009 (47) -15.44 (26.00) 46 -34.74 (46.94) 38
Houweling et al, 2011 (46) -3.86 (39.30) 102 -1.93 (29.77) 104
MacMahon et al, 2009 (48) -26.64 (37.45) 94 -6.17 (37.45) 94
New etal, 2003 (51) 189.20 (41.20) 345 200.01 (41.20) 338
Taylor et al, 2003 (32) -20.60 (26.00) 61 -11.50 (29.00) 66
Wallymahmed et al, 2011 (54)
Summary U2 = 90.8%)
166.00 (38.60) 40 200.80 (38.60) 41
Weighted Mean
Difference
(95% Cl), mg/dL
-3.00 (-14.14 to 8.14)
1.15 (-1.54 to 3.84)
-24.33 (-30.82 to -17.84)
19.30 (2.59 to 36.01)
-1.93 (-11.47 to 7.61)
-20.47 (-31.18 to -9.76)
-10.81 (-16.99 to -4.63)
-9.10 (-18.67 to 0.47)
-34.80 (-51.61 to -17.99)
-9.37 (-20.77 to 2.02)

-----1-----
- 4 0 - 2 0 0 2 0
Weighted Mean Difference, mg/dL
Total, n
Mean (SD) Mean (SD)
Allison et al, 1999 (39) -21.00 (31.00) 80 -23.00 (30.00) 72 I
DeBusk etal, 1994 (43) 106.95 (26.64) 243 131.66 (34.75) 244
■ •
Houweling et al, 2009 (47) -11.58 (26.03) 46 -23.17 (30.51) 38
MacMahon et al, 2009 (48) -20.85 (37.45) 94 -0.39 (37.45) 94 I-
----------- ■-------- 1
Taylor etal, 2003 (32) -19.40 (31.00) 61 -6.50 (30.00) 66 I------
--■—
Wallymahmed et al, 2011 (54) 84.94 (30.89) 40 111.97 (30.89)
41 I- -------- ■---------- 1
Summary (I2 = 89.1%)
- 4 5 - 2 5 0 2 5
Weighted Mean Difference, mg/dL
Weighted Mean
Difference
(95% Cl), mg/dL
2.00 (-7.70 to 11.70)
-24.71 (-30.21 t o -19.21)
11.59 (-0.69 to 23.87)
-20.46 (-31.17 t o -9.75)

-12.90 (-23.53 to -2.27)
-27.03 (-40.49 to -13.57)
-12.07 (-28.27 to 4.13)
To convert mg/dL to mmol/L, multiply by 0.0259.
1.02 to 2.31]), with substantial variability in treatment
effects (Q = 71.59; / 2 = 86%). Using the summary odds
ratio and median event rate from the control group of the
RCTs, we estimated the absolute treatment effect as a risk
difference o f 106 more patients achieving target total cho-
lesterol levels per 1000 patients (Cl, 5 to 196). Funnel
plots did not suggest publication bias (Supplement 6).
P a tie n t A d h e re n c e to T r e a tm e n t
Behavioral adherence was reported in 4 studies (39,
43, 48, 49). In 1 study, the rate o f daily medication adher-
ence (±SE) for the intervention group during the …
Disparities in Diabetes: The Nexus of Race, Poverty,
and Place
Darrell J. Gaskin, PhD, Roland J. Thorpe Jr, PhD, Emma E.
McGinty, PhD, MS, Kelly Bower, RN, PhD, Charles Rohde,
PhD,
J. Hunter Young, MD, MHS, Thomas A. LaVeist, PhD, and Lisa
Dubay, PhD, ScM
In the United States, 25.6 million or 11.3% of
adults aged 20 years and older had diabetes in
2010.1 Non-Hispanic Blacks had the highest
prevalence at 12.6% compared with non-
Hispanic Whites at 7.1%.1 Traditional expla-
nations for the observed race disparity in
diabetes prevalence include differences in

health behaviors, socioeconomic factors, family
history of diabetes, biological factors, and
environmental factors.2---4 Little work has been
conducted to understand how individual and
environment-level factors operate together to
produce disparities in diabetes prevalence.
A relatively new line of research has begun
to show that risk of diabetes is associated with
neighborhood attributes that are also associ-
ated with race. Auchincloss et al. found that
higher diabetes rates were related to lack of
availability of neighborhood resources that
support physical activity and healthy nutri-
tion.5 Schootman et al. found that poor housing
conditions were associated with diabetes prev-
alence.6 Black neighborhoods are more likely
to be characterized by these risk factors
(i.e., having food deserts, being less likely to
have recreational facilities, and tending to have
lower-quality housing than White neighbor-
hoods).7---18 As such it stands to reason that
failing to adjust national estimates of diabetes
prevalence for these social conditions might
influence perceptions of diabetes disparities.
LaVeist et al. compared disparities in diabetes
in an urban, racially integrated, low-income
community with a national sample from the
National Health Interview Survey.19,20 They
found that when urban Whites and Blacks
resided in the same low-income community,
the race disparity in diabetes prevalence dis-
appeared, largely because the prevalence rate
for Whites increased substantially.19 Ludwig
et al. used data from the Moving to Opportunity
demonstration project and found a lower

prevalence of diabetes among low-income
adults who moved from high-poverty
neighborhoods to low-poverty neighborhoods
compared with low-income adults who moved
from a high-poverty neighborhood to another
high-poverty neighborhood.21 Findings from
these studies suggest the need to further ex-
plore the role of place in race disparities in
diabetes.
We explored whether the nexus of race,
poverty, and neighborhood racial composition
and poverty concentration illuminates the race
disparities in diabetes. Specifically, we exam-
ined (1) whether diabetes prevalence increases
in predominantly Black neighborhoods com-
pared with predominantly White neighbor-
hoods, (2) whether diabetes prevalence is
higher in poor neighborhoods than in nonpoor
neighborhoods, and (3) whether the impact
of neighborhood racial composition and pov-
erty concentration on the risk of diabetes varies
by race. We hypothesized that residential
segregation and concentrated poverty (1) in-
crease Black individuals’ exposure to environ-
mental risks associated with poor health, (2)
reduce their access to community amenities
that promote good health and healthy behaviors,
and (3) limit their access to social determinants
that promote good health such as quality jobs,
education, public safety, and social net-
works.7,22---24
METHODS

The National Health and Nutrition Exami-
nation Survey (NHANES) was designed to de-
termine the health, functional, and nutritional
status of the US population. Since 1999,
NHANES has been conducted as a continuous,
annual survey with public use data files re-
leased in 2-year increments. Each sequential
series of this cross-sectional survey is a nation-
ally representative sample of the civilian non-
institutionalized population that consists of
an oversample of participants aged 12 to 19
years, participants aged 60 years and older,
Mexican Americans, Blacks, and low-income
individuals.25 Each of these surveys used
a stratified, multistage probability sampling
design.25 Data were collected from respon-
dents in 2 phases. The first phase consisted
of a home interview in which information
Objectives. We sought to determine the role of neighborhood
poverty and
racial composition on race disparities in diabetes prevalence.
Methods. We used data from the 1999–2004 National Health
and Nutrition
Examination Survey and 2000 US Census to estimate the impact
of individual
race and poverty and neighborhood racial composition and
poverty concentra-
tion on the odds of having diabetes.

Results. We found a race–poverty–place gradient for diabetes
prevalence for
Blacks and poor Whites. The odds of having diabetes were
higher for Blacks than
for Whites. Individual poverty increased the odds of having
diabetes for both
Whites and Blacks. Living in a poor neighborhood increased the
odds of having
diabetes for Blacks and poor Whites.
Conclusions. To address race disparities in diabetes,
policymakers should
address problems created by concentrated poverty (e.g., lack of
access to
reasonably priced fruits and vegetables, recreational facilities,
and health care
services; high crime rates; and greater exposures to
environmental toxins).
Housing and development policies in urban areas should avoid
creating high-
poverty neighborhoods. (Am J Public Health. 2014;104:2147–
2155. doi:10.2105/
AJPH.2013.301420)
RESEARCH AND PRACTICE

November 2014, Vol 104, No. 11 | American Journal of Public
Health Gaskin et al. | Peer Reviewed | Research and Practice |
2147
regarding the participant’s health history,
health behaviors, health utilization, and risk
factors were obtained. The second phase was
a medical examination. At the conclusion of the
home interview participants were invited to
receive a detailed physical examination at
a mobile examination center.25 Among those
who participated in the physical examination,
a nationally representative subset underwent
laboratory tests, including measurement of
fasting glucose.
We linked the NHANES data to 2000 US
Census data26 to measure the residential seg-
regation and concentrated poverty within
respondents’ census tract of residence. Because
we accessed the respondents’ census tract
information, the analysis was performed at the
National Center for Health Statistics (NCHS)
Research Data Center under the supervision
of NCHS staff to preserve the privacy, confi-
dentiality, and anonymity of the NHANES
respondents. In this analysis we used the
combined 1999---2004 data sets of adults who
completed the household interview, physical
examination, and laboratory components. We
restricted the analysis to Blacks (n = 1202) and
non-Hispanic Whites (n = 3201) who were
aged 25 years and older.

Key Dependent Variable and Independent
Variables
We identified persons with diabetes as re-
spondents who had a fasting glucose of 126
milligrams per deciliter or higher, had hemo-
globin A1c values of 6.5% or higher, or
reported taking medications for diabetes. We
excluded persons with normal glycemic values
who reported taking metformin from this
definition. Independent variables of interest
were individual race, individual poverty status,
neighborhood racial composition, and neigh-
borhood poverty concentration. Race was self-
reported in the NHANES as either non-Hispanic
African American/Black or non-Hispanic
White. We measured poverty status 2 ways.
The poverty---income ratio is a ratio of house-
hold income to the federal poverty level (FPL)
and is based on the respondent’s household
income and size.27 Poverty---income ratio was
coded as a 5-level categorical variable that
indicates each individual’s household poverty
ratio (below 100% of FPL [poor], 100% to
199% of FPL (near-poor), 200% to 299% of
FPL, 300% to 399% of FPL and greater than or
equal to 400% of FPL). We used this categori-
zation in our race---place model. Also, we used
a binary poverty variable indicating whether
individuals had household incomes between
0% and 199% of FPL or greater than or equal
to 200% of the FPL in our poverty---place
model.

We used the respondent’s census tract to
measure neighborhood characteristics because
census tracts are small, permanent, statistical
subdivisions within a county that range from
1500 to 8000 persons who are similar with
respect to population characteristics, economic
status, and living conditions. We designated
neighborhood racial composition as predomi-
nantly White, Black, or other race (Asian or
Hispanic) if that group was greater than 65% of
the census tract’s population. We designated
the racial composition of a neighborhood as
integrated if at least 2 groups were each more
that 35% of the census tract’s population. We
classified neighborhoods as having concen-
trated poverty if greater than or equal to 20%
of families in the census tract had incomes
below the FPL.
Other covariates included demographic
variables (age and gender), socioeconomic fac-
tors (education and health insurance status),
and family history of diabetes. We measured
age as a continuous variable. We included age
and age squared to control for nonlinearities.
We coded gender as a dichotomous variable.
We coded educational attainment as 5 cate-
gories (< 9 years of school, 9 to 12 years of
school but no diploma, high-school graduate or
general equivalency diploma, some college, or
college graduate or higher). We coded health
insurance coverage as 4 categories (privately
insured, Medicare, Medicaid or other govern-
ment coverage, or uninsured). We also con-
trolled for self-reported family history of
diabetes, if the respondent had any biological

relatives (grandparents, parents, brothers, or
sisters) who had been told by a health pro-
fessional that they had diabetes.
Statistical Analysis
We conducted bivariate analysis comparing
the diabetes prevalence across the categories
for each of our main independent variables.
We used the 2-by-N v2 test to determine
proportional differences by diabetes status. We
estimated a series of logistic regression models
to assess the intersection between diabetes
disparities and individual race and poverty and
neighborhood racial composition and poverty
concentration. The base model included all of
our key independent variables and covariates.
The race---place model interacted individual
race with neighborhood racial composition. To
do this, we created a variable with 8 categories:
White in White neighborhood, White in Black
neighborhood, White in other race neighbor-
hood, White in integrated neighborhood, Black
in Black neighborhood, Black in White neigh-
borhood, Black in other race neighborhood,
and Black in integrated neighborhood.
The poverty---place model combined indi-
vidual poverty with neighborhood poverty. We
created a variable with 4 categories: nonpoor
in nonpoor neighborhood, poor in nonpoor
neighborhood, nonpoor in poor neighborhood,
and poor in poor neighborhood. The race---
poverty---place model combined individual race
and poverty with neighborhood poverty. We

created a variable with 8 categories: nonpoor
White in nonpoor neighborhood, nonpoor
White in poor neighborhood, poor White in
nonpoor neighborhood, poor White in poor
neighborhood, nonpoor Black in nonpoor
neighborhood, nonpoor Black in poor neigh-
borhood, poor Black in nonpoor neighbor-
hood, and poor Black in poor neighborhood.
The sampling design for the NHANES is
a complex, stratified, multistage probability
sample of noninstitutionalized individuals.
Therefore, we developed sample weights to
account for both the differential probability of
being sampled and differential response rates.
We applied sample weights to account for the
differential probability of being selected, non-
response adjustments, and adjustments to na-
tional control totals in the NHANES.28
We adjusted parameter estimates and stan-
dard errors for the multistage sampling design
with Taylor linearization methods. Following
the algorithm described by the NCHS,29 we
created a 6-year sample weight variable by
assigning two thirds of the 4-year weight for
1999---2002 if the person was sampled in
1999---2002 or assigning one third of the
2-year weight for 2003---2004 if the person
was sampled in 2003---2004. We used the SVY
commands in Stata version 12 (StataCorp LP,
College Station, TX) to produce nationally
2148 | Research and Practice | Peer Reviewed | Gaskin et al.

American Journal of Public Health | November 2014, Vol 104,
No. 11
representative estimates and appropriate stan-
dard errors for all estimation.
RESULTS
The prevalence of diabetes varied with the
key independent variables and covariates
(Table 1). Blacks had a higher rate of diabetes
than Whites (0.123 vs 0.084; P = .03). The
prevalence of diabetes was inversely related to
household poverty level. Adults in poor and
near-poor households had the highest rates of
diabetes (0.12 and 0.127), followed by adults
between 200% and 299% FPL (0.108), fol-
lowed by adults between 300% and 399%
FPL (0.087), followed by adults in households
greater than or equal to 400% FPL (0.054).
Adults in predominantly Black neighborhoods
had higher rates of diabetes than those in
predominantly White neighborhoods (0.13 vs
0.084; P = .019). This neighborhood difference
is similar to the individual race difference.
When we combined individual race with
neighborhood racial composition, we found
that Blacks living in Black neighborhoods,
Blacks living in integrated neighborhoods, and
Blacks living in White neighborhoods had
significantly higher rates of diabetes (0.134,
0.123, and 0.106) than Whites in White
neighborhoods (0.083). When we combined

individual poverty with neighborhood poverty
concentration, we found that, compared with
nonpoor adults in nonpoor neighborhoods,
poor adults in poor and nonpoor neighbor-
hoods had higher rates of diabetes. When we
categorized adults by their race, poverty status,
and neighborhood poverty concentration, we
found that individual and neighborhood pov-
erty status were associated with diabetes for
Blacks and Whites.
Nonpoor Whites had lower rates of diabetes
than Blacks and poor Whites. Nonpoor Whites
in poor and nonpoor neighborhoods had sim-
ilar diabetes rates. There was a place gradient
for poor Whites. Poor Whites in poor neigh-
borhoods had the highest diabetes rates (0.15),
but the diabetes rate was lower for poor Whites
in nonpoor neighborhoods (0.121). For Blacks
there appears to be a race---poverty---place
gradient with nonpoor Blacks in nonpoor
neighborhoods having the lowest rates of di-
abetes (0.100), followed by poor Blacks in
nonpoor neighborhoods (0.114), nonpoor
TABLE 1—Diabetes Prevalence by the Independent Variables:
1999–2004 National Health
and Nutrition Examination Survey and 2000 US Census
Diabetes
Independent Variables No. Mean (95% CI) P
Individual race .03

Black 2605 0.123 (0.103, 0.144)
White 7184 0.084 (0.072, 0.958)
Individual poverty
Household poverty ‡400% FPL (Ref) 2989 0.053 (0.036, 0.071)
Household poverty 300%–399% FPL 1135 0.087 (0.059, 0.116)
.014
.017
<.001
Household poverty below FPL 1165 0.121 (0.0.87, 0.156) .004
Neighborhood poverty .037
Neighborhood concentrated poverty 2083 0.116 (0.089, 0.143)
Neighborhood no concentrated poverty 7701 0.084 (0.072,
0.096)
Neighborhood racial composition
Predominantly White neighborhood (Ref) 6668 0.084 (0.071,
0.097)
Predominantly Black neighborhood 1236 0.130 (0.101, 0.159)
.005
Predominantly other race neighborhood 200 0.119 (0.036,
0.020) .418

Integrated neighborhood 1680 0.094 (0.063, 0.124) .559
Race–place individual race and neighborhood racial
composition
White in White neighborhood (Ref) 6114 0.083 (0.070, 0.096)
White in Black neighborhood 42 0.072 (0.000, 0.216) .874
White in other race neighborhood 128 0.123 (0.021, 0.224) .451
White in integrated neighborhood 895 0.083 (0.046, 0.121) .994
Black in Black neighborhood 1194 0.134 (0.104, 0.165) .002
Black in White neighborhood 554 0.106 (0.059, 0.153) .0258
Black in other race neighborhood 72 0.108 (0.000, 0.223) .681
Black in integrated neighborhood 785 0.123 (0.083, 0.164) .048
Poverty–place individual poverty and neighborhood poverty
concentration
Nonpoor in nonpoor neighborhood (Ref) 4866 0.701 (0.058,
0.082)
Poor in nonpoor neighborhood 2149 0.120 (0.095, 0.145) <.001
Nonpoor in poor neighborhood 760 0.089 (0.048, 0.130) .339
Poor in poor neighborhood 1109 0.140 (0.010, 0.179) .003
Race–place–poverty individual race and poverty and
neighborhood

poverty concentration
Nonpoor White in nonpoor neighborhood (Ref) 4119 0.068
(0.056, 0.080)
Nonpoor White in poor neighborhood 275 0.062 (0.014, 0.111)
.828
Poor White in nonpoor neighborhood 1743 0.121 (0.095, 0.147)
<.001
Poor White in poor neighborhood 350 0.150 (0.071, 0.219) .043
Nonpoor Black in nonpoor neighborhood 667 0.100 (0.061,
0.141) .125
Nonpoor Black in poor neighborhood 485 0.136 (0.074, 0.198)
.030
Poor Black in nonpoor neighborhood 406 0.114 (0.057, 0.170)
.132
Poor Black in poor neighborhood 759 0.129 (0.129, 0.083) .011
Gender <.001
Male 5137 0.069 (0.058, 0.080)
Female 4652 0.110 (0.091, 0.129)
Continued

2149
Blacks in poor neighborhoods (0.136), and
then poor Blacks in poor neighborhoods
(0.129).
The base model determined if individual
covariates and neighborhood racial composi-
tion and poverty concentration separately in-
fluence the odds of having diabetes (Table 2).
We found that only household poverty status,
gender, and family history were significant
predictors. Neighborhood racial composition
and poverty concentration did not indepen-
dently influence the odds of having diabetes.
Compared with adults living at greater than or
equal to 400% FPL, the odds of having di-
abetes were 1.93 (95% confidence interval
[CI] = 1.21, 3.07) for the near-poor and 1.93
(95% CI = 1.09, 3.45) for the poor. The odds
of males having diabetes were 2.02 (95% CI =
1.59, 2.56) compared with females. The odds
of having diabetes among those with a family
history of diabetes were 3.27 (95% CI = 2.54,
4.21) compared with those without a family
history of diabetes.
The results from the race---place models
tested whether the odds of having diabetes
were related to adults’ racial identity relative to
the racial composition of their neighborhood
(Table 2). In this model, individual poverty
status, gender, and family history were still

significant predictors and similar in magnitude
to the base model; however, only Blacks in
integrated neighborhoods had greater odds
of having diabetes than Whites in White
neighborhoods (OR = 2.13; 95% CI = 1.26,
3.60). The other race---place indicator variables
were statistically insignificant.
The results from the poverty---place models
tested whether odds of having diabetes were
related to adults’ poverty status relative to their
neighborhood’s poverty concentration (Table 3).
We found that poor adults in nonpoor and
poor neighborhoods had greater odds of hav-
ing diabetes than nonpoor adults in nonpoor
neighborhoods. The odds of having diabetes
for poor adults in poor neighborhoods were
higher than for poor adults in nonpoor neigh-
borhoods (1.98 vs 1.67). Also, individual race
was significant in this model. The odds of
having diabetes were 1.59 (95% CI = 1.11,
2.28) times greater for Blacks than for Whites.
Finally, in the race---poverty---place model,
we categorized adults by their individual race,
individual poverty status, and neighborhood
poverty concentration. Similar to the bivariate
analysis, we found evidence of a race---poverty---
place gradient for poor Whites and nonpoor
Blacks in the logistic analysis. We found that,
compared with nonpoor Whites in nonpoor
neighborhoods, poor Whites in poor
TABLE 1—Continued

Family history of diabetes <.001
History of diabetes 4600 0.122 (0.103, 0.142)
No history of diabetes 5137 0.054 (0.043, 0.065)
Educational attainment
< 9th grade 775 0.195 (0.130, 0.259) .067
9th–12th grade, no diploma 1547 0.124 (0.090, 0.159) .006
High-school graduate (Ref) 2559 0.091 (0.071, 0.111)
Some college 2611 0.088 (0.068, 0.108) .077
‡ college graduate 2265 0.054 (0.032, 0.076) .002
Health insurance status
Private insurance (Ref) 6212 0.077 (0.065, 0.090)
Medicare 1702 0.200 (0.153, 0.248) <.001
Medicaid, SCHIP, or other government insurance 572 0.098
(0.060, 0.133) .569
No insurance 1303 0.054 (0.033, 0.075) .005
Note. CI = confidence interval; FPL = federal poverty level;
SCHIP = state children’s health insurance program.
TABLE 2—Estimated Odds Ratios of Having Diabetes by Race,
Concentrated Poverty, and
Racial Composition of Neighborhood: 1999–2004 National
Health and Nutrition

Examination Survey and 2000 US Census
Variable Base Model, OR (95% CI) Race–Place Model, OR
(95% CI)
Individual race
White (Ref) 1.00 . . .
Black 1.63 (0.94, 2.83) . . .
Concentrated poverty
Nonpoor neighborhood (Ref) 1.00 1.00
Poor neighborhood 1.02 (0.45, 1.93) 1.13 (0.75, 1.72)
Neighborhood racial composition
Predominantly White neighborhood (Ref) 1.00 . . .
Predominantly Black neighborhood 0.93 (0.45, 1.93) . . .
Predominantly other race neighborhood 1.16 (0.63, 2.14) . . .
Integrated neighborhood 1.30 (0.90, 1.88) . . .
Race–place individual race and neighborhood
racial composition
White in White neighborhood (Ref) . . . 1.00
White in Black neighborhood . . . 1.70 (0.24, 11.87)

White in other race neighborhood . . . 1.32 (0.34, 5.11)
White in integrated neighborhood . . . 1.32 (0.78, 2.24)
Black in Black neighborhood . . . 1.44 (0.92, 2.25)
Black in White neighborhood . . . 1.78 (0.87, 3.66)
Black in other race neighborhood . . . 1.30 (0.31, 5.55)
Black in integrated neighborhood . . . 2.13** (1.26, 3.60)
Continued
2150 | Research and Practice | Peer Reviewed | Gaskin et al.
American Journal of Public Health | November 2014, Vol 104,
No. 11
neighborhoods were the most disadvantaged
(OR = 2.51; 95% CI = 1.31, 4.81). The size of
the disadvantage was smaller for poor Whites
in nonpoor neighborhoods (OR = 1.73; 95%
CI = 1.16, 2.57). Compared with nonpoor
Whites in nonpoor neighborhoods, poor Blacks
in poor neighborhoods and nonpoor Blacks in
poor neighborhoods were similarly disadvan-
taged (OR = 2.45; 95% CI = 1.50, 4.01; and
OR = 2.49; 95% CI = 1.48, 4.19, respectively).
The size of the disadvantage was slightly lower
for poor Blacks in nonpoor neighborhoods
(OR = 2.34; 95% CI = 1.22, 4.46), and lower
for nonpoor Blacks in poor neighborhoods

(OR = 2.08; 95% CI = 1.26, 3.44). Although
the CIs overlap, the overall trends suggest that
there is a place gradient for poor Whites and
Blacks.
We estimated the predicted diabetes preva-
lence for the race---poverty---place categories
with adjustment for age, gender, socioeconomic
status, and diabetes family history (Figure 1).
We found that, for Whites, diabetes prevalence
was associated with individual poverty status,
and for poor Whites, neighborhood poverty
was associated with higher risk. For Blacks,
diabetes risk was associated with individual
and neighborhood poverty status ranging from
6.2% to 8.9%. However, neighborhood pov-
erty had a stronger association with diabetes
risk for nonpoor Blacks.
DISCUSSION
This study provides evidence that place
matters for Blacks and poor Whites. Living in
high-poverty neighborhoods increases the odds
of having diabetes for Blacks and poor Whites
but not for nonpoor Whites. Blacks and poor
Whites have higher odds of diabetes than
nonpoor Whites; however, living in poor
neighborhoods increases their odds further
such that poor Whites living in poor neigh-
borhoods are most disadvantaged. Our findings
are consistent with those of the Moving to
Opportunity demonstration project, which
demonstrated that enabling families to move

from high-poverty neighborhoods to low-
poverty neighborhoods improved their lives
along several dimensions, including general
health status, mental status, obesity rates, and
diabetes rates.21 Findings from a long-term
follow-up survey showed that Moving to
Opportunity participants who relocated to
low-poverty neighborhoods experienced
a 26% reduction in glycated hemoglobin level
of 6.5% or higher.30 A possible cause for this
reduction was changes in eating habits to
include more fruits and vegetables and an
increase in the amount of exercise.30
Why does living in a poor neighborhood
increase the odds of having diabetes for Blacks
and poor Whites? A recent report issued by
the Joint Center for Political and Economic
Studies showed that 46% of urban Blacks and
67% of poor urban Blacks live in high-poverty
neighborhoods (poverty rate > 20%) com-
pared with 11% of urban Whites and 30% of
poor urban Whites.31 The Exploring Health
Disparities in Integrated Communities study
reported that when poor Blacks and Whites
live in an integrated poor community, they
have similar diabetes prevalence (10.4% vs
10.5%).20 The narrowing of the disparities was
attributable to the White residents of this poor
community having higher rates of diabetes.
Other analyses of the Exploring Health Dis-
parities in Integrated Communities data found
similar results for obesity, hypertension, and
use of health services.19 The authors concluded
that community-level social and environmental
factors contribute to national race disparities

in diabetes. However, there are relatively few
integrated and economically balanced census
tracts in the United States (425 out of 66 438
in 2000). Concentrated poverty is not as large
a problem for Whites as it is for Blacks. Poor
Whites typically do not live in poor neighbor-
hoods. Black poverty is more concentrated
than White poverty; hence, poor Blacks have
greater exposure to negative neighborhood-
level health risks.
Poor Black neighborhoods may contribute
to higher diabetes prevalence because of the
decreased availability of healthy food and
limited walkability. These neighborhoods are
often referred to as “food deserts” because of
limited access to a supermarket or large gro-
cery store. Poor Black neighborhoods are more
TABLE 2—Continued
Individual poverty
Household poverty ‡400% (Ref) 1.00 1.00
Household poverty 300%–399% FPL 1.44 (0.92, 2.28) 1.56
(0.96, 2.53)
Household poverty 200%–299% FPL 1.48 (0.93, 2.37) 1.65*
(1.01, 2.68)
Household poverty 100%–199% FPL 1.93** (1.21, 3.07) 2.19**
(1.33, 3.61)
Household poverty below FPL 1.93* (1.09, 3.45) 2.35** (1.26,
4.40)

Gender
Female (Ref) 1.00 1.00
Male 2.02*** (1.59, 2.56) 2.17*** (1.64, 2.86)
Family history of diabetes
No family history of diabetes (Ref) 1.00 1.00
Family history of diabetes 3.27*** (2.54, 4.21) 2.94*** (2.22,
3.88)
Educational attainment
< 9th grade 1.19 (0.79, 1.79) 1.01 (0.60, 1.70)
9th–12th grade, no diploma 1.08 (0.71, 1.64) 1.00 (0.63, 1.58)
High-school graduate (Ref) 1.00 1.00
Some college 1.12 (0.79, 1.57) 1.07 (0.75, 1.54)
‡ college graduate 0.64 (0.36, 1.13) 0.61 (0.33, 1.14)
Health insurance status
Private insurance (Ref) 1.00 1.00
Medicare 1.26 (0.92, 1.72) 1.29 (0.90, 1.84)
Medicaid, SCHIP, or other government insurance 1.05 (0.63,
1.77) 0.90 (0.51, 1.58)
No insurance 0.77 (0.51, 1.16) 0.65 (0.36, 1.17)
Note. CI = confidence interval; FPL = federal poverty level; OR

= odds ratio; SCHIP = state children’s health insurance
program. The models controlled for age and quadratic age,
which were significant predictors (P < .001).
*P < .05; **P < .01; ***P < .001.
2151
likely to be “food deserts.” One study in Detroit
found that poor Black neighborhoods were
farther from supermarkets than poor White
neighborhoods.8 Another study found that
chain supermarkets were half as likely to be
located in predominantly Black neighborhoods
than in predominantly White neighborhoods.9
Several studies found that food available in
low-income and minority communities was
more expensive and of a lower quality.10---16
Morland and Filomena found that a lower
proportion of stores in predominantly Black
neighborhoods carried fresh produce, except
for bananas, potatoes, okra, and yucca.17 Blacks
in poor neighborhoods consume fewer fruits
and vegetables than people in middle-income,
racially integrated neighborhoods.32 This is
important because consumption of leafy green
vegetables is associated with a 14% reduced

risk of type 2 diabetes.33 There is strong
evidence suggesting that the walkability of
neighborhoods is positively associated with
physical activity and walking behaviors of
adults.34 In addition, residents of highly walk-
able neighborhoods are less likely to be over-
weight or obese.34---36
We did not find strong associations be-
tween diabetes prevalence and an individual’s
racial identity and the neighborhood racial
composition. Likewise, we did not find strong
associations between diabetes and an indi-
vidual’s poverty status and the neighbor-
hood’s poverty rate. Although there was
evidence of an individual race effect, neigh-
borhood racial composition does not seem to
have an effect on the odds of having diabetes.
The higher rate of diabetes prevalence …
CLINICAL SCHOLARSHIP
Multi-Ethnic Minority Nurses’ Knowledge and Practice
of Genetics and Genomics
Bernice Coleman, PhD, ACNP-BC, FAHA, FAAN1, Kathleen A.
Calzone, PhD, RN, APNG, FAAN2, Jean Jenkins,
PhD, RN, FAAN3, Carmen Paniagua, EdD, MSN, CPC, ANP,
ACNP-BC, AGACNP-BC, APNG-BC, FAANP4,
Reynaldo Rivera, DNP, RN, NEA-BC5, Oi Saeng Hong, RN,
PhD, FAAN6, Ida Spruill, PhD, RN, LISW, FAAN7,
& Vence Bonham, JD8
1 Research Scientist II, Nursing Research and Development,
Nurse Practitioner, Heart Transplant and Mechanical Assist

Device Programs, Heart
Institute, Cedars Sinai Medical Center, Los Angeles, CA, USA
2 Senior Nurse Specialist, Research, National Institutes of
Health, National Cancer Institute, Center for Cancer Research,
Genetics Branch, Bethesda, MD,
USA
3 Clinical Advisor, National Institutes of Health, National
Human Genome Research Institute, Bethesda, MD, USA
4 Adult Acute Care Nurse Practitioner & Adult Gerontology
Acute Care Nurse Practitioner, Advanced Practice Nurse
Geneticist, Department of
Emergency Medicine, University of Arkansas for Medical
Sciences, College of Medicine, Little Rock, AR, USA
5 Director of Nursing Innovation, New York-Presbyterian
Hospital, New York, NY, USA
6 Professor, University of California at San Francisco, School
of Nursing, Community Health Systems, San Francisco, CA,
USA
7 Assistant Professor, Medical University of South Carolina,
College of Nursing, Carleston, SC, USA
8 Associate Investigator, Social and Behavioral Research
Branch, National Institutes of Health, National Human Genome
Research Institute, Bethesda,
MD, USA
Key words
Minority nurses, nursing, genetics, survey,
nursing practice
Correspondence
Dr. Bernice Coleman, Nursing Research and
Development, Cedars Sinai Medical Center,
8700 Beverly Blvd., Los Angeles, CA 90048.

E-mail: [email protected]
Accepted: February 20, 2014
doi: 10.1111/jnu.12083
Abstract
Purpose: Exploratory studies establishing how well nurses have
integrated
genomics into practice have demonstrated there remains
opportunity for ed-
ucation. However, little is known about educational gaps in
multi-ethnic mi-
nority nurse populations. The purpose of this study was to
determine minority
nurses’ beliefs, practices, and competency in integrating
genetics-genomics in-
formation into practice using an online survey tool.
Design: A cross-sectional survey with registered nurses (RNs)
from the partic-
ipating National Coalition of Ethnic Minority Organizations
(NCEMNA). Two
phases were used: Phase one had a sample of 27 nurses who
determined the
feasibility of an online approach to survey completion and need
for tool revi-
sion. Phase two was a main survey with 389 participants who
completed the
revised survey. The survey ascertained the genomic knowledge,
beliefs, and
practice of a sample of multi-ethnic minority nurses who were
members of
associations comprising the NCEMNA.
Methods: The survey was administered online. Descriptive
survey responses

were analyzed using frequencies and percentages. Categorical
responses in
which comparisons were analyzed used chi square tests.
Findings: About 40% of the respondents held a master’s degree
(39%) and
42% worked in direct patient care. The majority of respondents
(79%) re-
ported that education in genomics was important. Ninety-five
percent agreed
or strongly agreed that family health history could identify at-
risk families,
85% reported knowing how to complete a second- and third-
generation fam-
ily history, and 63% felt family history was important to
nursing. Conversely,
50% of the respondents felt that their understanding of the
genetics of com-
mon disease was fair or poor, supported by 54% incorrectly
reporting they
thought heart disease and diabetes are caused by a single gene
variant. Only
30% reported taking a genetics course since licensure, and 94%
reported in-
terest in learning more about genomics. Eighty-four percent
believed that their
Journal of Nursing Scholarship, 2014; 46:4, 235–244. 235
C© 2014 Sigma Theta Tau International
Genomic Nursing Practice Coleman et al.
ethnic minority nurses’ organizations should have a visible role
in genetics and
genomics in their communities.

Conclusions: Most respondents felt genomics is important to
integrate into
practice but demonstrated knowledge deficits. There was strong
interest in the
need for continuing education and the role of the ethnic
minority organiza-
tions in facilitating the continuing education efforts. This study
provides evi-
dence of the need for targeted genomic education to prepare
ethnic minority
nurses to better translate genetics and genomics into practice.
Clinical Relevance: Genomics is critical to the practice of all
nurses, most
especially family health history assessment and the genomics of
common com-
plex diseases. There is a great opportunity and interest to
address the genetic-
genomic knowledge deficits in the nursing workforce as a
strategy to impact
patient outcomes.
As the proliferation of knowledge and understanding of
genomics accelerates, it becomes clearer that understand-
ing heritability and its intersection with environment has
now become foundational to nursing science, theory, and
practice. Genetic and genomic literacy now distinguishes
all nursing professionals as state-of-the-art academicians,
researchers, and clinicians who will provide the best care
possible. We are emerging into an era whereupon nursing
assessments, interventions, and the promotion of well-
ness will only attain scientific merit with the translation
of genomic knowledge to practice. Health care increas-
ingly demands that the registered nurse (RN) use ge-
nomic information and technology when designing and
providing care to those concerned about health or dis-
ease. These expectations have direct implications for RN

preparatory curricula, as well as for the 2.9 million prac-
ticing nurses (U.S. Department of Health and Human
Services, Health Resources and Services Administration,
2010).
Complex diseases such as cardiovascular and heart dis-
ease, diabetes, and cancer have disproportionally affected
racial and ethnic minority populations (National Center
for Health Statistics, 2012). While genetics research ex-
plores single gene disorders, the scientific discoveries now
inclusive of genomics are beginning to illuminate all ge-
netic variation in the human genome and the environ-
mental influences on health outcomes for persons with
complex chronic diseases. A transformative change in the
genomic knowledge of disease pathophysiology has pro-
duced a knowledge gap for nurses. A previous study as-
sessed nurses’ knowledge of genomics integration into
practice (Calzone et al., 2012; Calzone, Jenkins, Culp,
Bonham, & Badzek, 2013); however, the study was not
representative of ethnic minority nurses. In fact, very lit-
tle is known about genomic knowledge gaps of minor-
ity nurses (Spruill, Coleman, & Collins-McNeil, 2009).
These findings support the need for further investigation
of multi-ethnic minority nurses’ knowledge and practice
of genetics and genomics.
Background
The National Coalition of Ethnic Minority Nurse Asso-
ciations (NCEMNA) was incorporated in 1998 as a uni-
fied voice in nursing for the elimination of health dispari-
ties for ethnic minority populations. This national nursing
collaboration represents 350,000 nurses and is composed
of five ethnic minority nursing organizations. Its member
organizations are:

� Asian American/Pacific Islander Nurses Association,
Inc. (AAPINA)
� National Alaska Native American Indian Nurses As-
sociation, Inc. (NANAINA)
� National Association of Hispanic Nurses, Inc.
(NAHN)
� National Black Nurses Association, Inc. (NBNA)
� Philippine Nurses Association of America, Inc.
(PNAA)
The goals of the NCEMNA focus on development of
a cadre of ethnic nurses reflecting the nation’s diver-
sity, advocating for cultural competence, and accessible
and affordable health care. This coalition of ethnic mi-
nority nurse organizations collectively supports the de-
velopment of professional and educational advancement
of ethnic nurses, and the education of consumers, health-
care professionals, and policy makers on health issues of
ethnic minority populations. The NCEMNA’s primary ob-
jective is to develop ethnic minority nurse leaders in ar-
eas of health policy, practice, education, and research.
Through this approach, the endorsement of best nursing
practice models inclusive of genetics-genomics, educa-
tion, and research to improve the health of minority pop-
ulations is paramount (NCEMNA, 2013). One of the first
236 Journal of Nursing Scholarship, 2014; 46:4, 235–244.

Coleman et al. Genomic Nursing Practice
initiatives that the NCEMNA undertook was implement-
ing strategies to increase minority nurse participation and
success in research careers at the doctoral level. An area
determined as a collective interest to the NCEMNA mem-
ber organizations was the need to improve the health
of the representative ethnic minority patient populations
through research. Given the anticipated emerging major-
ity of these minority populations, the NCEMNA member
organizations identified the need to increase minority fac-
ulty and doctorally prepared nurses conducting research
through mentorship. Nurses from the NCEMNA member
organizations received competitive grants to participate
in the mentorship program that culminated in a yearly
conference where genetic-genomic information was pre-
sented as a foundational contributor to common diseases
found in ethnic patient populations represented by the
NCEMNA member organizations.
Representatives from the National Human Genome
Research Institute (NHGRI) and the National Cancer
Institute (NCI) along with the primary investigator of
this current work have presented on genetics and ge-
nomics at the National NCEMNA conferences. The re-
sponse and interest in genomic topics led to the interest
in gathering baseline information from these representa-
tive nursing groups regarding how ethnic minority nurses
utilized genetic-genomic core competencies and informa-
tion in their practice. Fundamental to this undertaking
was the establishment and endorsement of the Essential
Nursing Competencies and Curricula Guidelines for Ge-
netics and Genomics in October 2006 and expanded in
2008, and an established strategic implementation plan
that focused on practicing nurses, regulatory oversight
of nursing practice, and academic preparation of nurses

(Consensus Panel on Genetic/Genomic Nursing Compe-
tencies, 2006, 2009).
Theoretical Framework
The theoretical framework guiding this study was
Rogers’ Diffusion of Innovations (DOI; Rogers, 2003).
This theory consists of four components: (a) the inno-
vation, which in this study is genomics; (b) dissemi-
nation communication channels; (c) time; and (d) the
social system, which in this study is the minority nurs-
ing community. Factors that influence diffusion of the
innovation are antecedents and consist of adopter char-
acteristics as well as their attitudes. Adopters in this
study are the minority nurses, and their characteristics
include their genomic competency. Attitudes are the un-
derlying beliefs the adopters hold about the innovation
(i.e., genomics).
Study Aims
The ultimate goal of this collaborative project was to
assure that in this genomic era of health care, ethnic
minority nurses are prepared to assure quality care in
a diverse population that has concerns/experiences with
health disparities. Study aims were approached in two
phases to allow for testing of the study instrument fol-
lowed by administration of the instrument in the target
population.
Phase One Pilot Test Aims
1.1. Establish the feasibility of an online survey method
of data collection.
1.2. Evaluate the degree of respondent burden and sur-

vey response rates to establish whether this method of
data collection would be adequate for future target pop-
ulation implementation.
Phase Two Aims
2.1. Determine minority nurses’ beliefs, practices, and
competency of integrating into practice genomic informa-
tion related to common multifactorial diseases.
2.2. Assess knowledge of human genetic variation and
the use of patient characteristics, including ethnicity, gen-
der, genes, and race in diagnostics, treatment, and referral
decisions.
Analysis of aim 2.2 will be reported in a subsequent
article.
The NCEMNA Board approved moving forward with
a plan to utilize the diverse expertise of the NCEMNA
communities to create a genetics-genomics initiative. The
NANAINA chose to abstain from participation in this re-
search. Representatives from NCEMNA were identified to
organize this initiative with representatives of the NHGRI
and NIH. This study was approved by the Cedars Sinai
Institutional Review Board as well as the NIH Office of
Human Subjects Research.
Materials and Methods
Instrument
The survey instrument used in this study was collab-
oratively developed by all investigators. Multiple tele-
phone meetings were held to identify the process and re-
quired survey content to benchmark the genetic-genomic

knowledge of nurses via a membership survey. The fi-
nal draft survey is a compilation of the following five
instruments, which have been combined, reviewed, and
pretested by the research team.
1. The knowledge, attitude, and interest of African
American nurses toward genetics (Spruill et al.,
2009).
2. Bonham and Sellers’ Genetic Variation Knowledge
Assessment Index (GKAI; Bonham, Sellers, & Wool-
ford, submitted for publication).
3. Bonham and Sellers’ Health Professionals Beliefs
about Race (HPBR) scale.
4. Bonham and Sellers’ Racial Attributes in Clinical
Evaluation (RACE) scale.
5. The Genetics and Genomics in Nursing Practice
(GGNPS; Calzone et al., 2012).
The first survey instrument, the knowledge, beliefs,
and practices of African American nurses of genetics, was
designed to assess the interest, knowledge, and practice of
genetics and genomics among African American Nurses.
At tool construction, both face validity and construct va-
lidity were obtained using a panel of experts to evaluate
the items of the tool to ensure the construct was cap-

tured (Spruill et al., 2009). The Cronbach α standardized
is 0.652 for this 21-item survey instrument.
The survey instrument used in this study also included
questions modified from a study with physicians to eval-
uate nurses’ knowledge of genetic variation using the
Genetic Variation Knowledge Assessment Index (GKAI).
The GKAI scores range from 0 to 6, mean 3.28 (SD =
1.17) and was found to be symmetric and unimodal.
To evaluate nurses’ utilization of race in clinical prac-
tice, questions from the exploratory Health Profession-
als Beliefs about Race (HPBR; HPBR-BD, α = 0.69, four
items, and HPBR-CD α = 0.61, three items) and Racial
Attributes in Clinical Evaluation (RACE) scales (α = 0.86,
seven items; Bonham et al., submitted for publication).
In addition to the instruments described in the preced-
ing paragraph, the survey utilized for this study included
questions from the GGNPS instrument (Calzone et al.,
2012; Jenkins, Woolford, Stevens, Kahn, & McBride,
2010). This survey tool is constructed to evaluate Rogers
DOI theoretical domains, including attitudes, receptivity,
confidence, competency, knowledge, decision, and adop-
tion. Instrument validation was performed using struc-
tural equation modeling, which confirmed that the in-
strument items aligned with the domains of the DOI
(Jenkins et al., 2010).
The final compiled study instrument included seven
sections assessing beliefs, knowledge, practice, use of race
or ethnicity, education, and demographics. There were
a total of 61 questions, including multiple choice, di-
chotomous (yes or no), and Likert scale questions. The
questions were consistent with the Essentials of Genetic
and Genomic Nursing Competencies and assessed fam-
ily history utilization as well as the genomics of com-

mon disease, which represent knowledge and practice
expected of all RNs irrespective of their role, level of aca-
demic training, or specialty in which they practice (Con-
sensus Panel on Genetic/Genomic Nursing Competen-
cies, 2009). The selection of family history as evidence of
practice integration was intentional because family his-
tory collection falls within the scope of practice of all RNs
and is not cost or technology dependent.
Data Collection
Phase One. The target population consisted of nurses
attending the March 2009 NCEMNA conference. Nurses
of all levels of academic preparation, role, and clinical
specialty were invited to participate in the online survey
methodology assessing genetic and genomic knowledge,
belief, and skills. The only member organization exclu-
sion was NANAINA per their request. Conference leaders
provided notice to the 125 participants about the pilot
testing study, inviting them to test the instrument online.
No individual nurses were approached. Rather, interested
conference attendees self-selected to participate.
During Phase One pilot testing, computers were made
available at the NCEMNA annual meeting. A researcher
was stationed by the computer with an access code to as-
sist with survey access. A target of 30 participants was
desired for the study pilot phase. Prior to participation,
each participant was informed of the study aims and pro-
vided his/her verbal consent. In addition, upon launching
the survey online, the participant also had a written con-
sent as part of the instructions prior to encountering any
survey questions.
Phase Two. The following NCEMNA Associations

chose to participate: AAPINA, NAHN, NBNA, and PNAA.
Recruitment of study participants was done through each
participating NCEMNA member association. A link to the
survey was posted on the NCEMNA website as well as
each participating NCEMNA member association website.
Recruitment consisted of email announcements to associ-
ation constituencies as well as notifications through asso-
ciation newsletters. The survey offered no incentives. The
survey was open for a total of 10 months, with slightly
varying start dates for each association.
Instructions for the survey included the phone num-
bers and email addresses of study investigators to contact
with any questions. Participants also received instructions
that the survey was voluntary, no identifying informa-
tion would be collected or stored, and they could skip any
question.
Eligibility was limited only to licensed RNs who ac-
cessed the online survey. Membership in an NCEMNA
participating association was not required. Inclusion and
exclusion criteria were the same for both Phase One and
Phase Two studies.
Survey data were collected using the online survey
tool SurveyMonkey (SurveyMonkey, Inc., Palo Alto, CA,
USA). The survey took approximately 20 min for com-
pletion and collected no personal identifying information.

All data were stored in a password-protected file that was
available only to study investigators.
Statistical Analysis
Data were analyzed using SAS 9.3 (SAS Institute Inc.,
Cary, NC, USA). The answers to all survey questions were
summarized using descriptive statistical techniques. Chi-
squared tests were used to assess the relationships be-
tween survey items with categorical responses. The level
of significance was α = 0.05, and all tests of statistical sig-
nificance were two tailed.
Results
Phase One
A total of 27 participants completed the online sur-
vey. Participants found the length of the survey to be
just right. On average, participants spent 23 min com-
pleting the survey. There were some technical problems
with obtaining online access that were remedied during
Phase One of the study. The majority agreed or strongly
agreed that the directions for survey completion were
adequate 70% (n = 16/23), the survey was organized
86% (n = 20/23), the survey was easy to navigate 69%
(n = 16/23), question sequence was clear and predictable
70% (n = 16/23), terminology was consistent and ap-
propriate 82% (n = 19/23), and the survey was tech-
nically easy to complete 78% (n = 18/23). Most (82%,
n = 18/22) indicated that there were no questions
worded in a way that were not sensitive to their ethnic
group. Survey tool modifications were made based on
recommendations from the participants to enhance re-
spondent response by decreasing the number of survey
items. The final instrument for use in Phase Two con-

sisted of seven sections and a total of 61 questions.
Phase Two
Demographic and work characteristics of par-
ticipants. A total of 392 respondents completed an
online survey located on their nursing organization’s
website in Phase Two of the study. Excluding three in-
eligible participants reporting a highest nursing degree of
a licensed practical nurse, a total of 389 were included
in the data analysis. Table 1 summarizes the characteris-
tics of the eligible nurses. Participants’ ages ranged from
Table 1. Demographic Characteristics of Study Participants
Demographics (N = 389) n (%)
Sex (n = 326)
Male 22 (7%)
Female 304 (93%)
Age (n = 261)
Mean (range) 52 (23–82)
Race (n = 322)
White 27 (8%)
Asian 138 (43%)
Black/African American 107 (33%)
American Indian/Alaska Native 2 (1%)
Native Hawaiian/Pacific Islander 9 (3%)

Other 39 (12%)
Hispanic/Latino (n = 329) 60 (18%)
Highest level of nursing education (n = 331)
Diploma 5 (2%)
Associate degree 28 (8%)
Baccalaureate degree 115 (35%)
Master’s degree 130 (39%)
Doctoral degree 53 (16%)
Primary role (n = 330)
Administration 63 (19%)
Education 71 (22%)
Research 20 (6%)
Patient care 139 (42%)
Other 37 (11%)
Percent of time spent seeing patients (n = 311)
Mean 51%
Range 0–100%
NCEMNA organization affiliation (n = 305)
Asian American/Pacific Islander Nurses Association 37 (12%)
National Association of Hispanic Nurses 53 (17%)

National Black Nurses Association 109 (36%)
Philippine Nurses Association of America 112 (37%)
23 to 82 years, with a mean of 52 years, the majority
were female (93%, n = 304/326). The majority of par-
ticipants were Asian (43%, n = 138/322) and African
American (33%, n = 107/322). Eighteen percent (n =
60/329) stated that they considered themselves to be His-
panic/Latino, and 8% (n = 27/322) reported that they
were White. The majority (39%, n = 130/331) reported
their highest level of education was a master’s degree,
35% (n = 115/331) had a baccalaureate degree, 16%
(n = 53/331) held a doctoral degree, 8% (n = 28/331)
had an associate degree, and 2% (n = 5/331) were
diploma prepared. The primary work setting reported was
a hospital (68%, n = 163/241). The average number of
years they had worked in nursing was 20 years, and more
than half (51%, n = 166/326) had worked at their cur-
rent work setting for over 10 years. Forty-two percent
(n = 139/330) indicated their primary role was patient
care, 22% (n = 71/330) were in education, and 19% (n =
63/330) were in administration.
Beliefs. The majority of respondents felt it was very
important (79%, n = 301/383) or somewhat important
(19%, n = 71/383) for nurses to become more educated
about the genomics of common disease. The most fre-
quent advantages of integrating genomics into practice

identified included better decisions about recommenda-
tions for preventive services (87%, n = 332/383), bet-
ter treatment decisions (73%, n = 280/383), improved
services to patients (68%, n = 259/383), better ad-
herence to clinical recommendations by patients (56%,
n = 216/383), and genetic risk triaging (46%, n =
177/383). The highest reported potential disadvantages to
integrating genomics into practice included that it would
increase insurance discrimination (61%, n = 224/366),
genetics could increase patient anxiety about risk (52%,
n = 191/366), and it would be not reimbursable or too
costly (49%, n = 181/366).
Knowledge. Self-reported genetic knowledge as-
sessments are provided in Table 2. Half of the partici-
pants (50%, n = 182/364) felt their understanding of the
genetics of common diseases was poor or fair. The ma-
jority (95%, n = 371/389) agreed or strongly agreed that
family history could help to identify at-risk families and
85% (n = 323/381) knew how to complete it. The major-
ity had completed a family history for themselves (74%,
n = 279/378) and 51% (n = 195/381) had collected one
for a family member.
Responses varied by disease as to the degree to which
nurses felt genetics had clinical relevance to a wide range
of common health conditions. For example, only 54%
(n = 191/353) reported that hemochromatosis, an inher-
ited condition, had a great deal to do with genetics. The
majority correctly identified that genetic risk (e.g., as indi-
cated by family history) has clinical relevance for breast,
colon, and ovarian cancers; coronary heart disease; and
diabetes. However, 54% of respondents (n = 105/193)
thought diabetes and heart disease are caused by a single
gene variant, which is incorrect.

Practice. When presented with the option to identify
what was important to consider when delivering nursing
care, genes (29%, n = 53/185) and insurance (10%, n =
37/362) were the two lowest items identified as essential.
Other items scored as more essential to consider included
race (52%, n = 196/376), gender (53%, n = 196/371),
age (63%, n = 231/369), and family history (63%, n =
238/375).
Seventy-two percent (n = 274/380) also reported
collecting family histories for patients in their prac-
tice setting. When a patient indicated a disorder in
the family, nurses always collected the age of diagno-
sis (64%, n = 231/361), the relationship to the patient
Table 2. Knowledge Measures
Measure n (%)
Understanding of genetics of common diseases (n = 364)
Excellent 6 (2%)
Very good 47 (13%)
Good 129 (35%)
Fair 149 (41%)
Poor 33 (9%)
Do you think that genetic risk (e.g., as indicated by family
health history) has clinical relevance for breast cancer?
(n = 378)
Correct 378 (100%)

Incorrect 0 (0%)
health history) has clinical relevance for colon cancer?
(n = 375)
Correct 366 (98%)
Incorrect 9 (2%)
health history) has clinical relevance for coronary heart
disease? (n = 372)
Correct 333 (98%)
Incorrect 9 (2%)
health history) has clinical relevance for diabetes? (n =
376)
Correct 372 (99%)
Incorrect 4 (1%)
health history) has clinical relevance for ovarian
cancer? (n = 369)
Correct 354 (96%)

Incorrect 15 (4%)
The DNA sequences of two randomly selected healthy
individuals of the same sex are 90%–95% identical. (n =
208)
Correct 82 (39%)
Incorrect 126 (61%)
Most common diseases such as diabetes and heart
disease are caused by a single gene variant. (n = 193)
Correct 88 (46%)
Incorrect 105 (54%)
Genetics course since licensure (n = 356)
Yes 123 (35%)
No 233 (65%)
(91%, n = 330/363), race or ethnic background (77%,
n = 242/315), age at death from the condition (65%,
n = 237/362), as well as maternal and paternal lineages
(77%, n = 278/359).
With regard to family history specific knowledge el-
ements, nurses with higher levels of education tended
to accurately report that a family history should include
age at diagnosis of condition (p = .0146). More years
of practice influenced the collection by nurses of stan-
dard family history information that also included race or

ethnic backgrounds (p = .0197), age at death from con-
ditions (p = .0268), and age at diagnosis of condition
(p = .0009). Most nurses (98%, n = 380/386) agreed
or strongly agreed that family health histories could
be used to teach patients and family members about
the importance of genetics-genomics and disease pre-
vention. However, there was no relationship between
the proportion of work time spent seeing patients and
the perceived value of family history, use of family his-
tory, or variable collected (i.e., age, relationship, race, or
lineages).
Genetics and genomics education. Only 35%
(n = 123/356) indicated that they had taken a course that
included genetics as a major component since they ob-
tained their nursing license. While the majority of nurses
(94%, n = 335/357) indicated that they intended to learn
more about genetics, only 30% (n = 107/352) knew
whether there were any courses on genetics available
to them. More than half (55%, n = 196/358) identi-
fied workshops that included a mixture of presentations
and group activities as the preferred format for learning
about genetics. Overall, most (90%, n = 318/354) would
encourage NCEMNA or their organization to support
a genetics and genomics awareness initiative and 81%
(n = 289/357) responded that they would attend train-
ing if offered at their annual conference. Similarly, 84%
(n = 297/354) believed that their national organization
should have a visible role in genetics-genomics in their

community.
Discussion
This study …
OPINION ARTICLE Open Access
Genomics is changing personal healthcare
and medicine: the dawn of iPH
(individualized preventive healthcare)
Ruty Mehrian-Shai1 and Juergen K. V. Reichardt2,3*
Abstract
This opinion piece focuses on the convergence of information
technology (IT) in the form of personal monitors, especially
smart phones and possibly also smart watches, individual
genomic information and preventive healthcare and medicine.
This may benefit each one of us not only individually but also
society as a whole through iPH (individualized preventive
healthcare). This shift driven by genomic and other technologies
may well also change the relationship between patient
and physician by empowering the former but giving him/her
also much more individual responsibility.
Keywords: Human genomics, Individual information,
Personalized medicine, Medical education, Health care cost
Costs for healthcare in most countries are rising rapidly
and account for a sizeable fraction of a country’s GDP
(gross domestic product) [1]. This trend is most evident
in the USA where the fraction of GDP spent on health-
care has doubled from 8.2 % in 1980 to 16.2 % in 2012

[1]. This generally rising trend is noticeable in Australia
as well [1], although it is not as pronounced with an in-
crease from 5.8 % of GDP in 1980 to 8.6 % in 2011.
Clearly, this escalation is not sustainable and hence can-
not continue indefinitely. Healthcare must be sustain-
able. In fact, a significant burden is expended towards
the end of life [2] suggesting that a more preventive ap-
proach may be beneficial.
We propose here that a convergence of information
technology epitomized by individual monitors, incl.
smart phones and smart watches, and genomics in the
form of personal genomic information, especially on dis-
ease susceptibility, will result in new health information
accessible to each individuum.
The four converging areas leading to what we propose
to call individualized preventive healthcare (iPH) are:
First, ongoing rapid advances in personal monitors,
e.g. monitoring heart rate or tracking day to day activity,
e.g. smart phones and smart watches allow individuals to
collect, monitor and collate relevant health information
personally. These data can then be analyzed through on-
line world-wide searches, e.g. “Googling”, by the patient
him/herself before seeing a physician. There are also
significant ethical issues associated with these new devel-
opments [3] which must be carefully considered and
addressed.
Furthermore, genome sequencing is now approaching
a cost of just $1000 [4]. This price, which is continu-
ously falling, will put one’s own whole human genome
DNA sequence and its information at individual finger-
tips. Clearly, such genomic disease-related risk informa-

tion must be accompanied by appropriate and careful
interpretation and counselling [5]. In any case, individual
genomic information can be used to identify risks which
can then be mitigated if not eliminated altogether. Of
course, these developments in genomic science again
put the patient at the very heart of the matter by allow-
ing him/her to search for information, e.g. by Googling,
before seeing a physician to prevent (or at least slow)
disease.
Third, the microbiome [6] which is intrinsically per-
sonal and largely determined genomically also has be-
come of considerable interest and will find its way into
modern medical practice, perhaps again by patients
Googling information. In fact, because of the significant
role of the gut microbiota in human physiology and
* Correspondence: [email protected]
2Division of Tropical Health and Medicine, James Cook
University, Townsville,
QLD, Australia
3Present Address: Yachay Tech University, San Miguel de
Urcuquí, Ecuador
Full list of author information is available at the end of the
article
© 2015 Mehrian-Shai and Reichardt. Open Access This article
is distributed under the terms of the Creative Commons
Attribution 4.0 International License
(http://creativecommons.org/licenses/by/4.0/), which permits
unrestricted use, distribution,
and reproduction in any medium, provided you give appropriate
credit to the original author(s) and the source, provide a link
to the Creative Commons license, and indicate if changes were
made. The Creative Commons Public Domain Dedication
waiver (http://creativecommons.org/publicdomain/zero/1.0/)

applies to the data made available in this article, unless
otherwise
stated.
Mehrian-Shai and Reichardt Human Genomics (2015) 9:29
DOI 10.1186/s40246-015-0052-0
http://crossmark.crossref.org/dialog/?doi=10.1186/s40246-015-
0052-0&domain=pdf
mailto:[email protected]
http://creativecommons.org/licenses/by/4.0/
http://creativecommons.org/publicdomain/zero/1.0/
disease [6], new and unique opportunities will arise for
personal control of the gut flora. This will result in novel
strategies to prevent and treat diseases including cancer,
inflammatory bowel disease (IBD), diabetes, heart dis-
ease, allergy and perhaps even mental illness. The patho-
genesis of disease can be influenced also by various
epigenomic factors: microbiota, food intake, stress level
and physical activity. All these factors can be monitored,
investigated and evaluated.
We also note that the US NIH/NCI initiative on per-
sonalized medicine [7] to accelerate precision medicine
and the plan to monitor genetic and environmental fac-
tors of “cohort” of 1 million or more Americans will set
the basis of the multifactorial disease “warning” machin-
ery and provide valuable new insights.
Lastly, there is also an urgent need for credible and
trusted sources of medical information on the Internet
for individual patients to access and inform themselves.
This important issue has been addressed already, e.g. [8],
but will require constant attention, especially from all of

us, the medical professionals. Similarly, relationships
with patients are apt to change if they “arm” themselves
with Googled information.
Conclusion
In conclusion, we believe that iPH (individualized prevent-
ive healthcare) which arises from a convergence of per-
sonal monitors, incl. information technology (IT),
genomics, incl. the microbiome and vastly expanded infor-
mation available online will offer not only great individual
benefits by improving health through personalized infor-
mation and prevention but also significant cost savings in
the long run for healthcare. Furthermore, iPH may radic-
ally alter the relationship between physicians and patients.
This will give patients not only increased information but
also significant individual responsibility. Future research,
education and thoughtful discourse should prepare indi-
viduals, medical practitioners, scientists, (health) econo-
mists if not societies at large for these important changes.
Abbreviations
GDP: gross domestic product; iPH: individualized preventive
healthcare;
IT: information technology.
Competing interests
There are no competing interests to declare.
Authors’ contributions
JKVR conceived and wrote the manuscript, whilst RMS
commented on it and
contributed ideas as well. Both authors read and approved the
final
manuscript.
Acknowledgement

JKVR gratefully acknowledges the opportunity to develop these
ideas at
James Cook University whilst also visiting the MedUni Vienna
and the TU
Dresden.
Author details
1Pediatric Hemato-Oncology, Chaim Sheba Medical Center,
Ramat Gan, Israel.
2Division of Tropical Health and Medicine, James Cook
University, Townsville,
QLD, Australia. 3Present Address: Yachay Tech University, San
Miguel de
Urcuquí, Ecuador.
Received: 29 September 2015 Accepted: 31 October 2015
References
1. Organization for co-operation and development stat extracts,
Health status.
2015. (Accessed at http://stats.oecd.org/
index.aspx?DataSetCode=HEALTH_STAT#)
2. Katelaris AG. Time to rethink end-of-life care. Med J Aust.
2011;194:563.
3. Mittelstadt B, Fairweather NB, McBride N, Shaw M. Ethical
issues of personal
health monitoring: a literature review. ETHICOMP 2011
Conference
Proceedings 2011.
4. Hayden EC. Is the $1,000 genome for real? Nature. 2014;10.
5. Ormond KE. From genetic counseling to “genomic
counseling”. Mol Genet

Genomic Med. 2013;1:189–93.
6. Hollister EB, Gao C, Versalovic J. Compositional and
functional features of
the gastrointestinal microbiome and their effects on human
health.
Gastroenterology. 2014;146:1449–58.
7. Collins FS, Varmus H. A new initiative on precision
medicine. N Engl J Med.
2015;372:793–5.
8. National Institues of Health, Evaluation Health Information
2015. (Accessed
at
http://www.nlm.nih.gov/medlineplus/evaluatinghealthinformatio
n.html).
Submit your next manuscript to BioMed Central
and take full advantage of:
• Convenient online submission
• Thorough peer review
• No space constraints or color figure charges
• Immediate publication on acceptance
• Inclusion in PubMed, CAS, Scopus and Google Scholar
• Research which is freely available for redistribution
Submit your manuscript at
www.biomedcentral.com/submit

Mehrian-Shai and Reichardt Human Genomics (2015) 9:29
Page 2 of 2
http://stats.oecd.org/index.aspx?DataSetCode=HEALTH_STAT
http://stats.oecd.org/index.aspx?DataSetCode=HEALTH_STAT
http://www.nlm.nih.gov/medlineplus/evaluatinghealthinformatio
n.html
BioMed Central publishes under the Creative Commons
Attribution License (CCAL). Under
the CCAL, authors retain copyright to the article but users are
allowed to download, reprint,
distribute and /or copy articles in BioMed Central journals, as
long as the original work is
properly cited.
https://www.nursingworld.org/practice-policy/nursing-
excellence/ethics/genetics/
March-April 2016 • Vol. 25/No. 2 91
Alexandra Plavskin, MS, RN, is Clinical Instructor, Hunter
College, New York, NY.
Genetics and Genomics of Pathogens:
Fighting Infections with Genome-
Sequencing Technology
G enetics is “the study ofheredity” (World HealthOrganization
[WHO], 2002,

para. 1), while genomics is defined
as “the study of genes and their
functions, and related techniques”
(para. 2). An expanded definition of
genomics indicates “genetics scruti-
nizes the functioning and composi-
tion of the single gene whereas
genomics addresses all genes and
their interrelationships in order to
identify their combined influence
on the growth and development of
the organism” (WHO, n.d., para. 3).
Population genetics explores trait
changes in a population and poten-
tial contributing factors (Gillespie,
2010). Phylogenetics is the study of
evolutionary relatedness between
organisms (Wiley & Lieberman,
2011).
Background
The study of human genetics and
genomics is imperative because the
leading causes of mortality in the
United States all have a genetic
component, including cancer, heart
disease, and diabetes (Calzone et al.,
2010). However, the study of genet-
ics and genomics of pathogens also
can have substantial impact on clin-
ical practice. The study of patho -
gens can help identify sources of
infection and manage outbreaks of
health care-associated infections
(HAIs), one of the top 10 causes of

ReviewE ffe c ts o f N u rs e -M a n a g e d P ro to c o.docx

ReviewE ffe c ts o f N u rs e -M a n a g e d P ro to c o.docx

Recommended

Recommended

More Related Content

Similar to ReviewE ffe c ts o f N u rs e -M a n a g e d P ro to c o.docx

Similar to ReviewE ffe c ts o f N u rs e -M a n a g e d P ro to c o.docx (20)

More from malbert5

More from malbert5 (20)

Recently uploaded

Recently uploaded (20)

ReviewE ffe c ts o f N u rs e -M a n a g e d P ro to c o.docx