The effect of a literacy training program on family medicine residents.

Abstract

Pediatric literacy promotion programs carried out in the primary care setting, such as Reach Out and Read (ROR), have been associated with improved language skills for preschool children. Primary care physicians have frequent contact with young families and may be well situated for a literacy promotion program for both children and adults. We examined whether introducing ROR and an adult literacy intervention improves family medicine residents' literacy knowledge, attitudes, and practices.
We conducted a single group pretest/posttest evaluation design study of residents in a family medicine residency program serving low-income families. Residents completed self-administered questionnaires assessing literacy knowledge, attitudes, and practice. Then, through educational conferences, precepting, and ROR, residents were trained to assess and counsel patients about literacy. The same questionnaire was readministered 8 months later.
All 24 (100%) residents completed both the pre- and post-intervention questionnaires. Literacy knowledge mean scores increased from 74.5% to 83.1%. After the intervention, residents reported a greater sense of comfort in counseling about childhood and adult literacy. After the intervention, a greater proportion of residents reported usually or always asking about literacy milestones (30.2% to 79.2%) and parent-child reading (65.2% to 97.8%) during well-child visits.
A family literacy promotion program improved family medicine residents' self-reported literacy knowledge, attitudes, and practices. Such interventions can be incorporated into the education of family medicine residents with meaningful results.

Full text

Page 1

582
September 2004
Family Medicine
Low literacy is a common problem among both chil-
dren and adults. Thirty-five percent of children who
enter kindergarten lack language skills that are prereq-
uisites of literacy acquisition.1 Children living in pov-
erty bear a disproportionate burden of language delay
and reading disability.2,3 An estimated 44 million Ameri-
can adults, 20% of the population, have poor functional
literacy skills.4 Poor literacy skills are associated with
unemployment, poverty, and high-risk health activities.5
Adult low literacy is associated with poor adherence to
medical regimens, increased rates of hospitalization,
and poor rates of diabetes control.6-9
Children whose parents frequently read to them are
more likely to learn to read at grade-appropriate lev-
els,4,10,11 but reading to a child at least a few times a
week is less common in parents with lower income and
lower education.11,12 Parents are more likely to read to
their child when their physician encourages them to do
so.13 Primary care physicians serving low-income fami-
lies are well positioned to address literacy problems by
regularly encouraging reading.
Reach Out and Read (ROR), a reading intervention
in primary care clinics, improves literacy activities for
low-income children. ROR consists of (1) volunteers
in clinic waiting rooms who model reading aloud to
young children, (2) physicians giving parents anticipa-
tory guidance about reading to young children, and (3)
physicians giving the child a new book during each well-
child visit between the ages of 6 months and 5 years.
After involvement with ROR, parents have 4 to 10 times
the odds of reading books with their children,13,14 and
children show clinically improved expressive and re-
ceptive language skills, important prerequisites for
learning to read.15-17
Hazzard et al studied pediatric residents before and
after a ROR-like literacy intervention and found a trend
toward increased literacy assessment of children,18 yet
no prior study has assessed the effect of an educational
The Effect of a Literacy Training Program
on Family Medicine Residents
Marjorie S. Rosenthal, MD; Mark J . Werner, MD; Norman H. Dubin, PhD
From the Robert Wood Johnson Clinical Scholars Program and the Divi-
sion of Community Pediatrics, University of North Carolina (Dr Rosenthal);
the Franklin Square Hospital Center, Baltimore (Dr Werner); and the MedStar
Research Institute, Baltimore (Dr Dubin).
Background and Objectives: Pediatric literacy promotion programs carried out in the primary care
setting, such as Reach Out and Read (ROR), have been associated with improved language skills for
preschool children. Primary care physicians have frequent contact with young families and may be well
situated for a literacy promotion program for both children and adults. We examined whether introduc-
ing ROR and an adult literacy intervention improves family medicine residents’ literacy knowledge,
attitudes, and practices. Methods: We conducted a single group pretest/posttest evaluation design study
of residents in a family medicine residency program serving low-income families. Residents completed
self-administered questionnaires assessing literacy knowledge, attitudes, and practice. Then, through
educational conferences, precepting, and ROR, residents were trained to assess and counsel patients
about literacy. The same questionnaire was readministered 8 months later. Results: All 24 (100%) resi-
dents completed both the pre- and post-intervention questionnaires. Literacy knowledge mean scores
increased from 74.5% to 83.1%. After the intervention, residents reported a greater sense of comfort in
counseling about childhood and adult literacy. After the intervention, a greater proportion of residents
reported usually or always asking about literacy milestones (30.2% to 79.2%) and parent-child reading
(65.2% to 97.8%) during well-child visits. Conclusions: A family literacy promotion program improved
family medicine residents’ self-reported literacy knowledge, attitudes, and practices. Such interventions
can be incorporated into the education of family medicine residents with meaningful results.
(Fam Med 2004;36(8):582-7.)

Page 2

583Vol. 36, No. 8
intervention focused on teaching residents to counsel
adults and children about literacy. We hypothesized that
an educational intervention would improve residents’
knowledge, attitudes, and practice concerning early
childhood and adult literacy.
Methods
We performed a single group pretest/posttest evalu-
ation of an educational intervention, using a repeated
measures design in which family medicine residents
were tested for knowledge, attitudes, and behaviors
about child and adult literacy before and after literacy
training. Over a 1-month period, using educational re-
sources from the ROR national office,19 one investiga-
tor and one community member taught a childhood and
adult literacy curriculum. Residents’ success in adopt-
ing the literacy curriculum to their practice was assessed
through a comparison of their pretest and posttest scores
on a confidential, self-administered questionnaire.
Setting
The Franklin Square Family Health Center (FHC)
serves as a primary care health center in Baltimore for
10,000 patients with 27,000 visits annually. There are
approximately 275 well-child visits per month by chil-
dren in the target age range of ROR (6 months to 5
years). The children seen in the FHC are 58% Cauca-
sian, 28% African American, and 3% Latino/Hispanic;
67% are insured through Medicaid or Medicaid HMO
and 15% are uninsured. In Baltimore County, where
the clinic is located, low literacy is highly prevalent.
Up to 30% of adults read below a ninth-grade level.20
Eight family medicine attending physicians, three
pediatric attending physicians, and 24 family medicine
resident physicians see patients for health maintenance
and acute care visits at the FHC. The FHC is operated
by the Department of Family Practice, which has an
affiliation with the University of Maryland Medical
School. The MedStar Institute Institutional Review
Board (Baltimore) granted approval for the study. Each
resident gave written consent prior to participation in
this study.
Instrument
To measure residents’ knowledge, attitudes, and prac-
tices, we adapted an instrument used by Hazzard et al.
This instrument was previously validated by experts in
pediatric literacy and found to be reliable in a previous
study of pediatric residents.18 The instrument surveys
knowledge of literacy counseling, attitudes toward lit-
eracy assessment and counseling during health main-
tenance visits, exposure to previous literacy training,
literacy assessment and counseling behaviors during
health maintenance visits, and perceived barriers to
these behaviors. We added five knowledge questions,
three attitude questions, and two practice questions to
also focus on adult literacy. We pilot tested all ques-
tions with faculty at the FHC, leading to changes on
two questions.
Intervention
After collecting baseline data, one investigator
trained all residents and attending physicians in literacy
assessment and guidance, using didactic sessions, role-
playing, and peer feedback, during one grand rounds
and two noon conferences. Additionally, a volunteer
trainer from a local adult literacy organization led one
grand rounds. The total time devoted to training was 4
hours.
Childhood literacy training included role-modeling
techniques in teaching parents developmentally appro-
priate literacy activities for each age group and dem-
onstrating such activities in the office visit through the
use of a new, developmentally appropriate book. One
investigator trained the residents to use the child’s book
during the well-child visit both to perform the neuro-
developmental exam (ie, watching a 1-year-old point
at a named object) and to observe the parent-child in-
teraction (ie, observing a parent watch her 18-month-
old child’s glee at holding the book and turning the
pages). Residents were advised that adding the literacy
component may increase the quality and quantity of
the information gathered.
Adult literacy training consisted of role-playing tech-
niques in nonjudgmental information-gathering and
counseling skills. For example, residents were taught
statements like, “Many people have trouble understand-
ing their physician’s instructions. Does this ever hap-
pen to you?” and “Instructions for immunizations are
often hard to read. Do you ever have trouble reading
them?” Clinicians were encouraged to give bookmarks
with the name and address of a free, local adult lit-
eracy organization. The bookmarks were also with the
child’s book and in the waiting room, exam rooms, and
bathrooms.
Volunteers spent most of their time reading to chil-
dren in the waiting room. Using public librarians as
trainers, each volunteer received training in the mis-
sion of ROR, child development, and advice on read-
ing aloud.
Assessment
Eight to 10 months after the literacy educational in-
tervention, each resident took a post-intervention sur-
vey that was identical to the pre-intervention survey.
The 8–10 month interval was chosen to allow short-
term changes to extinguish.
Data Analysis
We calculated a composite knowledge score (per-
cent correct of the knowledge questions) and the com-
posite attitude score, which was calculated from 5-point
Special Articles: Health Literacy and Family Medicine

Page 3

584
September 2004
Family Medicine
Likert scale items; higher scores represent more lit-
eracy-facilitative attitudes. Practice assessment was
based on two questions: the frequency with which resi-
dents (1) asked about reading or literacy milestones as
a part of their history or developmental assessment and
(2) discussed parent-child reading as a part of anticipa-
tory guidance. Residents were asked to answer ques-
tions about each of these two items for four children’s
age groups (less than 12 months, 12–24 months, 2–5
years. and greater than 5 years) using a 5-point Likert
scale. We coded the variable into never/rarely/some-
times versus usually/always and calculated the percent-
age of times that residents reported usually or always
performing the literacy-oriented task. Open-ended ques-
tions related to barriers to literacy counseling were
grouped by category.
Paired t tests were used to compare normally dis-
tributed variables before and after the intervention. The
Wilcoxon signed rank test was used for paired non-
parametric data. Logistic regression was used to assess
if improvements in practice were associated with ex-
posures, attitudes, or knowledge. Data were analyzed
using STATA Windows version 7.0 (College Station,
Tex) or Statgraphics Plus version 3 (Manugistics Inc,
Rockville, Md).
Results
All 24 (100%) family medicine residents completed
both questionnaires. Residents’ demographic charac-
teristics are shown in Table 1. Forty-six percent reported
reading to a child in the 6 months prior to the interven-
tion; 8% had received training in literacy counseling
prior to the intervention used in this study.
Pre-intervention Data
Average knowledge scores by year of residency
ranged from 70.6% for first-year residents to 77.3%
for second-year residents, with an overall average of
74.5% (Table 2). Residents’ knowledge was best in ar-
eas related to other streams of development. For ex-
ample, 96% knew the age a child turns pages in a board
book, a skill that depends on knowledge of fine motor
skills development as much as language development.
Residents’ knowledge was poorest in questions about
adult literacy.
Regarding attitudes (Table 3), residents reported, on
a 5-point Likert scale, that spending time during well-
child visits assessing literacy was worthwhile
(mean=4.8), and using a book would facilitate the as-
sessment (mean=4.2). Residents reported feeling less
comfortable discussing parental literacy than parental
use of illicit drugs (mean=2.2). Regarding their prac-
tices, residents reported usually/always asking about
reading or literacy milestones in 30.2% of well-child
visits and discussing parent-child reading 65.2% of the
time (Table 4).
Post-intervention Data
Mean knowledge scores increased significantly
(P<.001) after the intervention from 74.5% to 83.1%
(Table 2). The knowledge score improved in 67% of
residents. After the intervention, residents reported a
greater (P<0.001) sense of comfort in assessing literacy
in preschool and adult patients (Table 3), as well as in
counseling about childhood literacy and adult literacy.
The composite attitude scores increased from 87 to 93
(P<.001).
As noted, residents were exposed to the literacy in-
tervention through attendance at pediatric or family
medicine grand rounds on literacy (96%), reading ar-
ticles about literacy in primary care (74%), and read-
ing to a child during the intervention (52%). In addi-
tion, outside of the formal intervention, 52% discussed
the topic with an attending, and 13% observed an at-
tending model the intervention. No individual expo-
sure, nor group of exposures, was associated with in-
creased knowledge, increased literacy-oriented atti-
tudes, or increased literacy practice.
Self-reported literacy assessment improved with the
intervention (Table 4). “Usually or always” asking about
reading or literacy milestones increased from 30.2% to
79.2% (P<.001). “Usually or always” asking about par-
ent-child reading increased from 65.2% to 97.8%
(P<.001). Incorporating literacy into practice improved
for 92% of residents, and the increase was significant
within each year of residency. Knowledge, exposure,
and attitude post-intervention scores or change in scores
were not associated with practice outcome.
Table 1
Characteristics of Family Medicine Resident
Study Participants at Baseline (n=24)
Characteristic
Gender
Female
Male
n (%) or Mean
18 (75%)
6 (25%)
Race
Caucasian
African American
Asian
17 (71%)
1 (4%)
6 (25%)
Age30.5 years
Year in training
First
Second
Third
8 (33%)
7 (29%)
9 (38%)
Had read to child of his/her own in prior 6 months5 (21%)
Had read to a child in prior 6 months 11 (46%)
Had literacy training prior to the intervention2 (8%)

Page 4

585Vol. 36, No. 8
The residents reported that their biggest barriers to
literacy assessment and counseling in children were
time (67%) and knowledge (8%). In adults, the biggest
barriers were time (54%); trying to avoid an uncom-
fortable, awkward, or embarrassing situation (29%);
knowledge (17%); and inexperience (13%).
Discussion
Our data show that introducing a clinic-based ROR
and an adult literacy intervention improves family medi-
cine residents’ self-reported knowledge, attitudes, and
practice for literacy assessment of both children and
adults. Our findings extend prior research demonstrat-
ing the utility of counseling families about early child-
hood literacy activities.14-18 The increasing evidence of
the role of adult literacy on childhood health, combined
with the unique role of the family physician to care for
the whole family, argue for the importance of both as-
pects of the intervention.22,23
Our finding builds on and contrasts with a previous
study that failed to detect a statistically significant dif-
ference in pediatric residents, despite using a similar
intervention and similar instrument.18 Perhaps our in-
cluding strategies for both children and adults reinforced
the importance of literacy.
When asked about barriers to literacy assessment and
counseling in adults, 30% of residents reported being
uncomfortable or feared offending the parent; no resi-
dent mentioned this as a barrier to assessing children.
Recognizing that a parent’s literacy is associated with
a child’s literacy activities14 as well as parents’ percep-
tion of how sick a child is,22 adult literacy is an impor-
tant childhood issue. Since parents of young children
may not have their own regular source of health care
but regularly interact with their child’s physician, the
only person counseling adults on health-related topics
may be the child’s physician.23 The residents’ discom-
fort may represent an important educational barrier. Not
surprisingly, time was the most frequently reported
Special Articles: Health Literacy and Family Medicine
Table 2
Family Medicine Residents’ Knowledge
of Preschool and Adult Literacy
Mean % Correct (SD) *
Before
Intervention
All residents (n=24) 74.5(10.5)
First-year residents (n=8) 70.6(10.4)
Second-year residents (n=7) 77.3 (7.1)
Third-year residents (n=9)75.8(12.6)
After
Intervention
83.1 (8.9)
77.9 (9.3)
84.0 (8.1)
86.9 (7.7)
P Value
<0.001
0.13
0.04
0.01
SD—standard deviation
* Data were compared using two-sided paired t tests.
Table 3
Family Medicine Residents’ Attitudes Toward Assessing and Counseling About Literacy in Well-child Visits
Mean (SD)*
Before
Intervention
After
Intervention
P Value Range
“A child’s primary health care provider has a unique opportunity to encourage
behaviors that improve a child’s chances to succeed in reading.” 4.8 (0.4)4.8 (0.5) 1.0 1–5**
“Giving away a book during the well-child visit would facilitate talking with parents
about literacy.”4.2 (1.1) 4.7 (0.5).03 1–5
“I am more comfortable asking about literacy than illicit drug use.”2.2 (1.1)2.5 (1.2) .461–5
“I feel comfortable assessing literacy in my adult patients and parents
of my pediatric patients.” 2.3 (0.6) 3.0 (0.7).0011–5
“I feel comfortable assessing literacy in my preschool-age patients.” 2.6 (1.0)3.7 (0.7) <.0011–5
“I feel comfortable giving literacy-related counseling to my adult patients.”2.5 (0.6)3.4 (0.6)<.001 1–5
“I feel comfortable giving literacy-related counseling to parents concerning
their young children.”3.3 (0.9)4.0 (0.7)<.001 1–5
Overall literacy-oriented attitude score87 (5)93 (8)<.001 23–115
SD—standard deviation
*
** 1=strongly disagree, 5=strongly agree. A higher number indicates a more literacy-oriented attitude.
Data compared using two-sided paired t tests

Page 5

586
September 2004
Family Medicine
barrier. Clinicians who have incorporated ROR into
their practice, however, have anecdotally reported that
it improves the neuro-developmental exam and obser-
vation of the parent-child relationship without length-
ening the office visit. They also report improved bond-
ing among clinician, child, and parent.24
Limitations
There are several limitations to our study. First, the
study was conducted at a single site and had a small
sample size. Second, we had no control group. Thus,
we cannot determine whether observed improvements
were related to a secular trend or to the natural growth
that educators expect of residents throughout their train-
ing. Third, we cannot determine which components of
our intervention—didactic sessions, role-modeling,
precepting—were most important in changing behav-
iors. Fourth, we did not collect data on the dose of the
intervention (eg, number of residents who attended the
noon conferences, number of books each resident gave
to a child, number of well-child visits within the target
age range, number of families each resident counseled).
Finally, our outcomes were based on physician self-
report. They may have been providing socially accept-
able answers, although this would not explain observed
improvements over time. While a Hawthorne effect is
possible, we attempted to minimize this threat to valid-
ity by referring to our study as “Improving Parent and
Child Activities in the Family Health Center,” rather
than giving it a literacy-based name. Other studies that
have examined language and literacy outcomes in chil-
dren, however, find that when clinicians claim to have
succeeded in counseling, improvement occurs.24
Conclusions
Encouraging families to read to young children is
important for school readiness, for parental satisfac-
tion, and for many aspects of a child’s develop-
ment.22,25-27 A limited intervention, such as that described
here, can be implemented with free educational re-
sources from the ROR national office (www.
reachoutandread.org), voluntary teaching from commu-
nity literacy organizations, and fiscal support for the
children’s books.
Our results are important because the intervention
can be easily incorporated into family medicine resi-
dents’ education. Family physicians are well situated
to intervene simultaneously on behalf of children and
adults. ROR may be the only well-child intervention
that has been shown to affect parents’ attitudes and be-
haviors, as well as a child’s language development.24
Our data demonstrate that after a limited ROR inter-
vention, residents are capable and willing to incorpo-
rate literacy advocacy of both children and adults into
well-child care.
Future studies should evaluate primary care-based
literacy programs using randomized controlled trials.
Such future studies should collect outcome data from
parents, as well as from physicians.
Acknowledgments: Dr Rosenthal is currently a Robert Wood Johnson Clinical
Scholar. She was supported during the writing of the manuscript by the
Health Resources and Service Administration, grant #5 T32 HP 14001-14.
Funding for this project was also provided by the MedStar Research Insti-
tute, Baltimore, grants # 2000-129 and # 2001-106.
This study was presented in part at the Pediatric Academic Societies
Meeting, May 2002, in Baltimore.
The authors thank the family medicine residents at Franklin Square Hos-
pital Center for participation in the study, co-fellows and faculty in the NRSA
and RWJCS programs, and Amal Murarka, MD.
Corresponding Author: Address correspondence to Dr Rosenthal, Univer-
sity of North Carolina, Robert Wood Johnson Clinical Scholars Program,
Division of Community Pediatrics, 5045 Old Clinic Building CB #7105,
Chapel Hill, NC 27599-7105. 919-966-3713. Fax: 919-843-9237.
marjorie_rosenthal@med.unc.edu.
REFERENCES
1.Carnegie Foundation. Ready to learn: a mandate for the nation. New
York: Carnegie Foundation, 1991.
National Center for Education Statistics. The condition of education.
Washington, DC: US Department of Education, Office of Educational
Research and Improvement, 1996.
Anastasiouw N, Hanes ML, Hanes M. Language and reading strategies
for poverty children. Baltimore: University Park Press, 1982.
Kirsch I, Jungeblut A, Jemkins L, Kolstat A. Adult literacy in America:
a first look at the results of the National Adult Literacy Survey. Wash-
ington, DC: National Center for Education, US Department of Educa-
tion, 1993.
Berlin G, Sum A. Toward a more perfect union: basic skills, poor fami-
lies, and our economic future. New York: Ford Foundation, Project on
Social Welfare and the American Future, 1988.
Weiss BD, Hart G, McGee DL, D’Estelle S. Health status of illiterate
adults: relation between literacy and health status among persons with
low literacy skills. J Am Board Fam Pract 1992;5(3):257-64.
Kalichman SC, Ramachandran B, Catz S. Adherence to combination
antiretroviral therapies in HIV patients of low health literacy. J Gen
Intern Med 1999;14(5):267-73.
2.
3.
4.
5.
6.
7.
Table 4
Family Medicine Residents’ Practice of Literacy-
related Tasks in Well-child Visits
% of Times Answered
“Usually or Always” (SD)*
Before
Intervention
After
Intervention
P Value
Frequency of asking about
reading or literacy milestones in
a well-child visit 30.2 (37.6)79.2 (32.7)<.001
Frequency of discussing parent-
child reading in a well-child visit65.2 (35.9)97.8(7.2)<.001
SD—standard deviation
* Data compared using Wilcoxon signed rank test