Content uploaded by Gail L Daumit
Author content
All content in this area was uploaded by Gail L Daumit on Mar 03, 2014
Content may be subject to copyright.
Systematic Review: The Value of the Periodic Health Evaluation
L. Ebony Boulware, MD, MPH; Spyridon Marinopoulos, MD, MBA; Karran A. Phillips, MD, MSc; Constance W. Hwang, MD;
Kenric Maynor, MD; Dan Merenstein, MD; Renee F. Wilson, MSc; George J. Barnes, BA; Eric B. Bass, MD, MPH;
Neil R. Powe, MD, MPH, MBA; and Gail L. Daumit, MD, MHS
Background: The periodic health evaluation (PHE) has been a
fundamental part of medical practice for decades despite a lack of
consensus on its value.
Purpose: To synthesize the evidence on benefits and harms of the
PHE.
Data Sources: Electronic searches of such databases as MEDLINE
and the Cochrane Library, review of reference lists, and hand-
searching of journals through September 2006.
Study Selection: Studies (English-language only) assessing the de-
livery of preventive services, clinical outcomes, and costs among
patients receiving the PHE versus those receiving usual care.
Data Extraction: Study design and settings, descriptions of the
PHE, and clinical outcomes associated with the PHE.
Data Synthesis: The best available evidence assessing benefits or
harms of the PHE consisted of 21 studies published from 1973 to
2004. The PHE had a consistently beneficial association with patient
receipt of gynecologic examinations and Papanicolaou smears, cho-
lesterol screening, and fecal occult blood testing. The PHE also had
a beneficial effect on patient “worry” in 1 randomized, controlled
trial but had mixed effects on other clinical outcomes and costs.
Limitations: Descriptions of the PHE and outcomes were hetero-
geneous. Some trials were performed before U.S. Preventive Ser-
vices Task Force guidelines were disseminated, limiting their appli-
cability to modern practice.
Conclusions: Evidence suggests that the PHE improves delivery of
some recommended preventive services and may lessen patient
worry. Although additional research is needed to clarify the long-
term benefits, harms, and costs of receiving the PHE, evidence of
benefits in this study justifies implementation of the PHE in clinical
practice.
Ann Intern Med. 2007;146:289-300. www.annals.org
For author affiliations, see end of text.
T
he periodic health evaluation (PHE) has been a funda-
mental part of medical practice for decades, despite a
lack of consensus regarding its value in health promotion
and disease prevention. The PHE consists of one or more
visits with a health care provider to assess patients’ overall
health and risk factors for preventable disease, and it results
in the delivery of clinical preventive services that are tai-
lored to a patient’s age, sex, and clinical risk factors and
laboratory testing (1). By promoting prevention and en-
hancing the patient–provider relationship, the PHE may
improve patient outcomes and the public’s health (2).
However, it could also induce unnecessary costs and pa-
tient harms by promoting the use of nonrecommended
services. Early studies of the PHE, performed before the
adoption of current preventive services guidelines, were
costly and demonstrated minimal improvement in clinical
outcomes (3, 4). Because of resulting concern over the
PHE’s value, some experts have advocated for episodic,
targeted delivery of preventive services in the context of
ongoing clinical care (5, 6). More recent clinical trials have
reported some benefits of the PHE (7–11).
Private and public health insurance coverage for pre-
ventive services in the United States has gradually increased
over time, although generally for one recommended service
at a time rather than for a comprehensive set of preventive
services (12). Recent legislation provides coverage for a
“welcome to Medicare visit” for new enrollees, incorporat-
ing a range of diagnostic and screening tests (13). Despite
this legislation and continued use of the PHE, there is a
lack of clear evidence demonstrating that the PHE im-
proves patient outcomes or reduces health care costs.
In light of conflicting opinions regarding the PHE’s
impact on health and health care costs and nonuniformity
on its implementation, we performed a systematic review
of the evidence to ascertain the PHE’s benefits and harms
with regard to patient outcomes and health care costs.
METHODS
Study Design
We performed a systematic review for the Evidence-
Based Practice Center Program of the Agency for Health-
care Research and Quality (AHRQ). A review of evidence
identifying the value of the PHE was initially nominated to
the AHRQ by the American College of Physicians. We
developed a conceptual model to help define the PHE and
its potential benefits and harms, identified relevant studies
that assessed benefits and harms of the PHE, extracted
data, assessed individual study quality, and synthesized the
evidence.
Conceptualization of the PHE
In the absence of standard definitions of the PHE, we
developed a conceptual framework to guide our assessment
of the value of the PHE by 1) identifying the potential
See also:
Web-Only
Appendix Table
Conversion of figures and tables into slides
Annals of Internal Medicine Review
© 2007 American College of Physicians 289
goals, benefits, and harms of the PHE and 2) clarifying
how the PHE might be consistently identified in the pub-
lished literature. In this framework, patient, provider, and
societal goals of the PHE could include the promotion of
personal and family health, patient education, and im-
provement of public health. The PHE itself could lead to
the delivery of appropriate (potential benefit) or inappro-
priate (potential harm) clinical preventive services and
could lead to other potential benefits and harms, such as
changes in patient attitudes and behaviors, improvement or
worsening of clinical outcomes, decreases or increases in
health system resource use and costs, and improvements or
decrements in public health. Both health care provider and
health system characteristics could modify the effect of
PHE on outcomes (Figure 1).
Definition of the PHE and Usual Care
Because the PHE is tailored to individual patients and
is thus delivered in a highly heterogeneous fashion on the
basis of clinical resources available to different health care
providers, we sought to develop a definition that could be
widely applied to a majority of clinical practice environ-
ments, regardless of patient populations, health care deliv-
ery settings, or resource constraints. We also used a prelim-
inary review of representative studies to help define the
PHE in a manner that would allow us to identify the
Figure 1. Conceptual framework developed to guide assessment of the value of the periodic health evaluation.
Pap ⫽ Papanicolaou.
Review Periodic Health Evaluation
290 20 February 2007 Annals of Internal Medicine Volume 146 • Number 4 www.annals.org
broadest selection of studies assessing its value. We defined
the PHE as one or more visits with a health care provider
for the primary purpose of assessing patients’ overall health
and risk factors for disease that may be prevented by early
intervention. Our definition specified the PHE as consist-
ing only of the history, risk assessment, and a tailored phys-
ical examination that could lead to the delivery of preven-
tive services. According to our definition, the PHE did not
include the delivery of clinical preventive services that pa-
tients could receive during or after their visit for the PHE
and that we considered an outcome of the PHE. For in-
stance, a 50-year-old woman receiving a PHE would
undergo a detailed history, risk assessment, and physical
examination (which could include a gynecologic examina-
tion). Under our definition, the delivery of clinical preven-
tive services provided both during that visit (such as coun-
seling to stop smoking and a Papanicolaou [Pap] smear)
and outside of the visit (such as mammography or colonos-
copy) were considered to be a result of the PHE (history,
risk assessment, and physical examination) and not part of
the PHE.
We defined “usual care” as the delivery of clinical pre-
ventive services in the absence of a health care provider
visit designated for the primary purpose of assessing pa-
Figure 2. Summary of literature search and review process (number of articles).
Excluded: 1284
*Total may exceed 1284; several reasons for exclusion at the abstract level were allowed. †A total of 50 articles were included in the data abstraction.
These 50 articles represented 33 studies that reported multiple outcomes and multiple follow-ups. ‡Total may exceed 769; several reasons for exclusion
at the article inclusion/exclusion review level were allowed. CINAHL ⫽ Cumulative Index of Nursing and Alliance Health Literature; PHE ⫽ periodic
health evaluation.
ReviewPeriodic Health Evaluation
www.annals.org 20 February 2007 Annals of Internal Medicine Volume 146 • Number 4 291
tients’ health and risk factors for disease. Under this defi-
nition, preventive services were considered to have been
delivered opportunistically (that is, in the setting of a
health care provider visit designated for the ongoing care of
chronic illnesses or other acute illnesses).
Identification of Relevant Studies
We searched MEDLINE, the Cochrane Libraries, the
Health Technology Assessment Database (HTA), the Na-
tional Health System Economic Evaluation Database
(NHS EED), and the Cumulative Index of Nursing and
Allied Health Literature (CINAHL) for studies published
through September 2006 that compared the PHE with
usual care and assessed benefits and harms of the PHE.
Examples of search terms included periodic physical examina-
tion, periodic health evaluation, annual physical examination,
annual check up, multiphasic screening, multiphasic health test-
ing, preventive screening, preventive services, and well care visits.
To identify studies that our search strategy might have missed,
we reviewed reference lists of relevant articles, and we hand-
searched tables of contents of 24 periodicals in general medi-
cine, preventive medicine, and public health. Titles and ab-
stracts deemed potentially relevant were further reviewed if
either of 2 reviewers did not exclude them. For articles pro-
moted to abstract review, 2 investigators independently re-
viewed abstracts and excluded them if they 1) had no useful
information applying to the benefits or harms of the PHE, 2)
Table 1. Components of Risk Assessment Described as Part of Periodic Health Evaluation in 33 Studies Included in Literature Review
Component of Risk Assessed Study, Year (Reference)
Observational Studies
Grimaldi,
1965 (22)
Roberts et al.,
1969 (23)
Slesinger et al.,
1976 (24)
Geiger et al.,
1993 (25)
Nakanishi et al.,
1996 (20)
Kottke et al.,
1997 (26)
Sox et al.,
1997 (27)
Bernacki et al.,
1998 (28)
Diet √
Physical activity
Substance abuse
Injury prevention
Safe sexual practice
Tobacco smoking √
Calcium intake
Folic acid
Sun exposure
Oral health
Medications/polypharmacy
Not otherwise specified √√ √ √ √ √
Table 2. Components of Physical Examination Described as Part of Periodic Health Evaluation in 33 Studies Included in Literature Review
Component of
Physical Examination
Study, Year (Reference)
Observational Studies
Grimaldi,
1965 (22)
Roberts et al.,
1969 (23)
Slesinger et al.,
1976 (24)
Geiger et al.,
1993 (25)
Nakanishi et al.,
1996 (20)
Kottke et al.,
1997 (26)
Sox et al.,
1997 (27)
Bernacki et al.,
1998 (28)
Blood pressure √√√
Height
Weight
Pulse
Cardiovascular
Pulmonary
Abdominal
Neurologic
Breast √√√
Gynecologic √√√
Rectal
Prostate √
Foot
Funduscopic
“Examination,” not
otherwise specified
√√ √ √ √ √
Other* √
* For example, vision testing, tonometry, and audiometry.
Review Periodic Health Evaluation
292 20 February 2007 Annals of Internal Medicine Volume 146 • Number 4 www.annals.org
were not written in English, 3) included participants 18 years
or younger, 4) contained no original data, or 5) had no com-
parison group. We included observational studies as well as
randomized, controlled trials (RCTs). Reviewers were paired
randomly at all stages.
Data Extraction
Two reviewers sequentially abstracted data for each
article, including information on the studies’ designs, loca-
tions and settings, dates of performance, follow-up length,
enrollment, eligibility criteria, participant characteristics,
components of the PHE, interventions, and outcomes.
Data were abstracted to capture changes in the delivery (by
health care providers) or receipt (by patients) of recom-
mended preventive services as a result of the PHE, including
the delivery of recommended aspects of the physical examina-
tion (such as blood pressure measurement and gynecologic
examination), counseling (such as substance abuse counsel-
ing), immunizations (such as influenza vaccination), and
screening tests (such as cholesterol testing). Data were also
abstracted regarding changes in patient attitudes and percep-
tions as a result of the PHE (such as knowledge and satisfac-
tion), patient behavioral outcomes (such as rates of tobacco
cessation), proximal or intermediate clinical outcomes (such as
cholesterol lowering and disease detection), distal clinical out-
comes (such as death), economic outcomes (such as cost and
Table 1—Continued
Study, Year (Reference)
Observational Studies
Williams et al.,
1998 (29)
Hahn,
1999 (30)
Freedman et al.,
2000 (31)
Stange et al.,
2000 (32)
Hama et al.,
2001 (33)
Nutting et al.,
2001 (34)
Parchman and Byrd,
2001 (35)
Tao et al.,
2001 (36)
Burton et al.,
2002 (37)
Chiou and Chang,
2002 (38)
√√ √
√ √
√√ √
√√ √
√ √
√ √
√
√
√ √
√ √√
√
√√ √√√ √
Continued on following page
Table 2—Continued
Study, Year (Reference)
Observational Studies
Williams et al.,
1998 (29)
Hahn,
1999 (30)
Freedman et al.,
2000 (31)
Stange et al.,
2000 (32)
Hama et al.,
2001 (33)
Nutting et al.,
2001 (34)
Parchman and Byrd,
2001 (35)
Tao et al.,
2001 (36)
Burton et al.,
2002 (37)
Chiou and Chang,
2002 (38)
√√√ √ √√√
√√ √√
√√ √√
√√
√
√√
√
√
√
√√√
√ √√
√√
√
√√
√√ √√ √√
√√√ √√√
Continued on following page
ReviewPeriodic Health Evaluation
www.annals.org 20 February 2007 Annals of Internal Medicine Volume 146 • Number 4 293
health care utilization), and public health outcomes (such as
containment of communicable disease).
Study Quality Assessment
Two reviewers independently judged each study’s
quality on several aspects of external and internal validity,
including descriptions of 1) inclusion and exclusion criteria
for participants, 2) participants’ baseline characteristics, 3)
nonenrollees, 4) handling of withdrawals, 5) the interven-
tion, 6) adequacy of length of follow-up, 7) participant
attrition, 8) outcomes, 9) relevancy and appropriateness of
outcomes, 10) quality of outcomes assessment, 11) quality
of randomization for RCTs, 12) quality of blinding for
RCTs, 13) similarities and differences in the management
of study groups for RCTs, 14) comparable characteristics
of enrolled participants for control and treatment groups
for RCTs, and 15) statistical analysis. For both experimen-
tal and observational studies, we applied a total quality
score based on work by Chalmers and colleagues (14).
Evidence Synthesis
We synthesized findings from multiple studies by as-
sessing the quantity, quality, and consistency of the “best
available evidence” on the benefits and harms of the PHE.
We adapted an evidence-grading scheme recommended by
the Grading of Recommendations, Assessment, Develop-
ment and Evaluation (GRADE) Working Group (15). The
GRADE classification scheme incorporates a systematic ap-
proach toward assessing the entire body of literature on
specific outcomes in which “points” are assigned to (or
Table 1—Continued
Study, Year (Reference)
Observational Studies Trials
Finkelstein,
2002 (39)
Schneider et al.,
2003 (40)
Flocke and Stange,
2004 (41)
Lin et al.,
2004 (42)
Somkin et al.,
2004 (43)
Cutler et al.,
1973 (6)
Fletcher et al.,
1977 (18)
Stone et al.,
1981 (5)
Belcher,
1990 (19)
Burton et al.,
1995 (10)
√√ √
√ √
√√ √√
√
√√ √
√√ √√
√
√
√ √
√
√
√√√√√√
Table 2—Continued
Study, Year (Reference)
Observational Studies Trials
Finkelstein,
2002 (39)
Schneider et al.,
2003 (40)
Flocke and Stange,
2004 (41)
Lin et al.,
2004 (42)
Somkin et al.,
2004 (43)
Cutler et al.,
1973 (6)
Fletcher et al.,
1977 (18)
Stone et al.,
1981 (5)
Belcher,
1990 (19)
Burton et al.,
1995 (10)
√√√√
√√ √
√ √√ √
√
√
√
√√
√√
√√ √
√√ √
√
√
√
√
√√ √ √√ √
√√√√
Review Periodic Health Evaluation
294 20 February 2007 Annals of Internal Medicine Volume 146 • Number 4 www.annals.org
subtracted from) scores for each body of evidence, based on
prespecified criteria, including assessments of individual
study quality, evaluation of the consistency of the direction
of results reported on specific outcomes, handling of plau-
sible confounders across all studies evaluating an outcome,
the strength of the associations between the PHE and out-
comes, and the directness of evidence linking the PHE to
outcomes across all studies. In assigning GRADE classifi-
cations, study members reviewed the evidence on each out-
come as a group, and final GRADE classifications were
arrived at by group consensus (16). We considered the best
available evidence for each outcome to consist of at least 2
RCTs or at least 2 observational studies with designs least
likely to present biased findings (cohort studies [considered
best], followed by cross-sectional studies and studies with
pre–post observational design [considered worst]). A
GRADE classification of “high” signified that further re-
search would be unlikely to alter our conclusions regarding
the association of the PHE with outcomes, “medium” sig-
nified that further research could alter our conclusions,
“low” signified that further research would be very likely to
alter our conclusions, and “very low” signified that further
research would alter our conclusions.
Assessing the Magnitude of Effect of the PHE on
Outcomes
To quantify the magnitude of the effect of the PHE
on outcomes in a standard way among studies reporting a
variety of heterogeneous outcomes (for example, percent-
age of persons receiving clinical preventive services in some
studies vs. mean changes in clinical measures [such as
blood pressure] in other studies), we calculated the Cohen
d effect size estimate (95% CI) for mean differences and
differences in proportions among all RCTs. We then clas-
sified effects as small, intermediate, or large effects by using
standard criteria (Appendix Table, available at www.annals
.org) (17). We considered evidence to show a clear benefi-
cial effect of the PHE when the investigators reported that
the PHE consistently resulted in greater benefits or a re-
duction in harms compared with usual care in all studies
assessing that outcome. We considered evidence to show a
clear harmful effect of the PHE when investigators re-
ported that the PHE consistently resulted in fewer benefits,
more harms, or a smaller reduction in harms compared
with usual care in all studies assessing that outcome. We
considered evidence to have no effect when findings were
consistently neutral (that is, the 95% CI of the estimate of
reported effects included 0). We considered evidence to
show a mixed effect of the PHE when investigators of some
studies assessing the PHE reported beneficial effects while
others reported harmful or no effects.
Role of the Funding Source
Technical experts from the funding source (AHRQ)
provided guidance regarding all aspects of the conduct of
the review.
RESULTS
Yield of Literature Search and Identified Studies
We screened 7039 articles for eligibility at the title
review level and reviewed 2103 at the abstract level. Of
these, 50 articles were eligible for full review, representing
33 studies (10 RCTs [5–11, 18, 19, 21] and 23 observa-
tional studies [20, 22– 43]) reporting on the benefits or
harms of the PHE (Figure 2).
Definitions of the Adult PHE in Studies of Its Value
Definitions of the PHE were heterogeneous within
studies. While central elements used to define the PHE in
studies included the clinical history and risk assessment of
patients as well as a physical examination, the specific com-
position of these central elements varied among studies.
Table 1—Continued
Study, Year (Reference)
Trials
Elder et al.,
1995 (8)
Morrissey et al.,
1995 (9)
Nakanishi,
1996 (20)
Theobald,
1998 (21)
Patrick,
1999 (7)
√√
√√ √
√√ √
√√
√√ √
√
√√√
Table 2—Continued
Study, Year (Reference)
Trials
Elder et al.,
1995 (8)
Morrissey et al.,
1995 (9)
Nakanishi,
1996 (20)
Theobald,
1998 (21)
Patrick,
1999 (7)
√√
√
√
√
√√
√√
√
√√
√√√√
√√
ReviewPeriodic Health Evaluation
www.annals.org 20 February 2007 Annals of Internal Medicine Volume 146 • Number 4 295
The most frequently cited types of history and risk assess-
ment performed were assessment of alcohol or substance
abuse, tobacco smoking, and dietary risks; the least fre-
quently cited types of risk assessment included assessment
of calcium and folic acid intake (Table 1). Reports of al-
most two thirds of studies mentioned the physical exami-
nation element of the PHE but did not identify the com-
ponents. In the remaining one third of studies, the most
frequently cited components were assessment of blood
pressure, weight, and height; breast examination; gyneco-
logic examination; and rectal examination. The least fre-
quently cited components included neurologic and foot
examinations (Table 2). With the exception of studies re-
porting on the joint delivery of the gynecologic examina-
tion or Pap smear as part of the definition of the PHE, no
studies included the delivery of clinical preventive services
(for example, counseling, immunization, or preventive test-
ing) as part of their description of the PHE.
Design and Setting of Identified Studies
Overall, the literature was characterized by complexity
and heterogeneity in several dimensions. Among the 33
eligible studies, 10 were RCTs, 8 were cohort studies (2
retrospective), 12 were cross-sectional, and 3 featured pre–
post comparisons of patients before and after undergoing a
PHE. Studies were conducted over a period of several de-
cades, with nearly one third performed before 1989. While
more than two thirds of studies were performed in the
United States, we also identified relevant studies from the
United Kingdom, Canada, Taiwan, Japan, Denmark, and
Sweden. Practice settings for the studies were diverse, with
studies taking place in private offices, academic practices,
hospital outpatient clinics, and other settings. Studies re-
flected a range of health care systems.
Best Available Evidence of Benefits and Harms of PHE
Compared with Care without a PHE
Ten RCTs, 2 cohort studies, and 9 cross-sectional
studies constituted the best available evidence on benefits
and harms of the PHE. These studies were published be-
tween 1973 and 2004 and reported on a wide range of
outcomes, including delivery of preventive services, proxi-
mal clinical outcomes, and distal clinical outcomes. Studies
not included among the best available evidence were ob-
servational studies (cohort studies, cross-sectional studies,
and studies with pre–post designs) deemed to contribute
more biased estimates than the best available evidence for
Table 3. Summary of Results from Best Available Evidence to Assess Each Outcome*
Outcome Type of Evidence Considered (Studies,
n
) GRADE Classification† Effect (Range of Magnitude)
of PHE on Outcome‡
Delivery of clinical preventive services
Gynecologic examination/
Papanicolaou smear
RCTs (2) High Beneficial (small to large)
Counseling RCTs (1)
Observational (6)
Low Mixed
Immunizations RCTs (3) Medium Mixed
Cholesterol screening RCTs (1)
Observational (4)
Medium Beneficial (small to large)
Colon cancer screening (fecal occult
blood testing)
RCTs (2) High Beneficial (large)
Mammography RCTs (1)
Observational (1)
Low Mixed
Proximal clinical outcomes
Disease detection RCTs (2) Medium Mixed
Health habits RCTs (5) Medium Mixed
Patient attitudes (worry) RCTs (1) Medium Beneficial§
Health status RCTs (2) Medium Mixed§
Blood pressure RCTs (2) High Mixed
Serum cholesterol RCTs (1)
Observational (1)
Low Mixed
Body mass index RCTs (3) Medium Mixed
Distal economic and clinical outcomes
Costs RCTs (4) Medium Mixed
Disability RCTs (2) Medium Mixed
Hospitalization RCTs (3) High Mixed
Mortality RCTs (5) Medium Mixed
* GRADE ⫽ Grading of Recommendations, Assessment, and Evaluation; Observational ⫽ studies with observational design; PHE ⫽ periodic health evaluation; RCT ⫽
randomized, controlled trial.
† Adapted from the Grading of Recommendations, Assessment, Development, and Evaluation (GRADE) Working Group.
‡ Magnitude and direction of effect of receipt of PHE on outcome, based on standardized effect sizes calculated by using the Cohen d statistic. We considered effect sizes
ranging from 0 to 0.25 to represent small effects, those ranging from 0.26 to 0.8 to represent intermediate effects, and those greater than 0.8 to represent large effects. Effect
sizes can be thought of as the average percentile standing of the average treated (or experimental) participant relative to the average untreated (or control) participant. An effect
size of 0.0 indicates that the mean of the treated group is at the 50th percentile of the untreated group. An effect size of 0.25 indicates that the mean of the treated group
is at the 58th percentile of the untreated group. An effect size of 0.8 indicates that the mean of the treated group is at the 79th percentile of the untreated group.
§ Standardized effect size could not be calculated for the study or some studies assessing this outcome.
Review Periodic Health Evaluation
296 20 February 2007 Annals of Internal Medicine Volume 146 • Number 4 www.annals.org
each outcome. The full evidence report (16) describes these
studies in detail.
Overall, the strength and consistency of the evidence
varied widely among outcomes, as did the magnitude and
direction of results. Nearly half of studies reported on mul-
tiple outcomes: Twelve studies reported on 1 outcome; 9
studies reported on 2 or more outcomes. The full evidence
report produced for AHRQ contains details of results from
the 21 studies that provided the best available evidence,
including scores for specific criteria and overall grades on
the strength and consistency of the evidence (16). Data on
effect size estimates can be viewed in the Appendix Table
(available at www.annals.org).
Summary of Results from the Best Available Evidence on
Benefits and Harms of the PHE
Delivery of Clinical Preventive Services
Four RCTs (7, 9, 19, 44) and 10 observational studies
(26, 27, 29, 30, 32, 35, 36, 39, 41, 42) assessed delivery of
clinical preventive services as a result of the PHE. The
overall GRADE classification of the evidence assessing
these outcomes ranged from low to high. Evidence on gy-
necologic examination or Pap smear and fecal occult blood
screening received a high rating, evidence on immuniza-
tions and cholesterol screening received a medium rating,
and evidence on counseling and mammography receiving a
low rating. The PHE had a beneficial association with re-
ceipt of gynecologic examinations or Pap smears, choles-
terol screening, and fecal occult blood testing (greater rates
of delivery of these clinical preventive services in persons
undergoing the PHE than in those not undergoing the
PHE) and had mixed effects on other clinical preventive
services (Table 3).
Proximal Clinical Outcomes
Six RCTs (3, 7, 8, 10, 11, 18, 44, 45) and 1 observa-
tional study (33) assessed the association of the PHE with
proximal clinical outcomes. Outcomes ranged from disease
detection to changes in patient health habits, attitudes, and
clinical measures (including blood pressure, serum choles-
terol, and body mass index) as a result of the PHE. The
overall GRADE classification of the evidence assessing
these outcomes ranged from low to high. Evidence on se-
rum cholesterol received a low rating; evidence on disease
detection, health habits, patient attitudes, health status,
and body mass index received a medium rating; and evi-
dence on blood pressure received a high rating. The PHE
had a beneficial effect on patient worry in 1 RCT (7) (less
increase in patient worry over time among persons under-
going the PHE than in those not undergoing the PHE),
but associations between the PHE and other proximal clin-
ical outcomes were mixed (Table 3).
Distal Economic and Clinical Outcomes
Five RCTs (3, 4, 9, 10, 21, 46 – 48) assessed distal
economic and clinical outcomes resulting from the PHE.
Outcomes included costs, disability, hospitalization, and
mortality. The overall GRADE classification of the evi-
dence assessing these outcomes ranged from medium to
high. Evidence on costs, disability, and mortality received a
medium rating, while evidence on hospitalization received
a high rating. The PHE was found to have mixed effects on
all distal economic and clinical outcomes (Table 3).
DISCUSSION
The best available evidence suggests that patients ben-
efit from the PHE through its association with improved
delivery of some recommended clinical preventive services
and through reduction of patient worry. The available ev-
idence does not reveal harms associated with the PHE.
Given that short- and long-term studies have shown that
appropriate implementation of currently recommended
preventive services improves health in short and long-term
studies (49) and that elimination of worry or concern re-
garding illness may represent a powerful motivator for ac-
tion on the part of patients (50 –56), our findings provide
health care providers and payers with justification for the
continued implementation of the PHE.
Mechanisms through which improvements in care at-
tributed to the PHE occur are unclear, as studies were
highly heterogeneous in terms of content of the PHE and
their institution of additional interventions to enhance de-
livery of the PHE. Regardless of the specific components
included, the PHE may provide clinicians time to consider
preventive care more fully, thus leading to their instituting
preventive measures more frequently (57). It is possible
that the PHE has a stronger effect on improving the deliv-
ery of preventive services that are performed by clinicians
at the time of the office visit (such as gynecologic exami-
nations and Pap smears) than on preventive services that
require patients to schedule appointments outside of the
initial office visit for the PHE, which might also be affected
by patient adherence or system failures (58). By providing
an opportunity for both patients and physicians to contem-
plate and discuss potential risks, the PHE could also pro-
vide a vehicle through which patient worries can be more
fully elucidated from patients and addressed. Findings of
clear benefits of the PHE despite great study heterogeneity
could provide clinicians with confidence that the PHE may
confer benefits in their own practices. Nevertheless, clini-
cians should use individual judgment regarding the opti-
mal way in which the PHE is implemented given the lo-
gistics in their practices and resource constraints.
Several gaps in the identified literature are notable and
lay the foundation for future research aimed at further
elucidating the value of the PHE. The PHE was delivered
heterogeneously and with varying levels of intensity in
studies. Such heterogeneity prohibited any determination
of what might constitute an adequate PHE to achieve ben-
efits. Randomized trials comparing the effect of different
characteristics of the PHE on clinical outcomes, including
ReviewPeriodic Health Evaluation
www.annals.org 20 February 2007 Annals of Internal Medicine Volume 146 • Number 4 297
variations in the frequency and intensity of specific com-
ponents of the PHE, are needed. Studies comparing the
effect of the PHE in varying patient populations and sys-
tems of care are also needed to elucidate who will best
benefit from the PHE. Evidence was mixed with regard to
most short- and long-term clinical outcomes reported in
the literature. Well-performed, long-term clinical trials are
needed to identify whether the PHE can consistently im-
prove intermediate (for example, patient attitudes, health
status serum cholesterol, blood pressure) and long-term
clinical outcomes (for example, hospitalization, costs, or
death) in contemporary medical settings, to characterize
the effect of the PHE on the patient–physician relation-
ship, and to assess the effect of the PHE on broad societal
outcomes, such as disease containment in populations.
Large-scale trials could be costly and potentially unable to
capture long-term effects of the PHE on such outcomes as
costs and mortality because of multiple competing factors
affecting these outcomes (for example, changes in medical
technologies and their demonstrated benefits over time);
however, the development of computerized models to sim-
ulate trajectories of quality of life, the development of mor-
bidity and mortality, and impacts on direct and indirect
costs as a result of the PHE are needed.
Limitations of this synthesis and the literature deserve
mention. First, although we based our definition of the
PHE on a conceptual model that provided context for as-
sessing its value to patients, providers, and society, our
definition was also based partially on our desire to broadly
identify studies assessing the value of the PHE. Thus, our
definition could be construed as allowing so much hetero-
geneity in the PHE among identified studies as to obscure
definitions that might be commonly used among clini-
cians. For instance, many health care providers may con-
sider the delivery of clinical preventive services to be part
of—and not a result of—the PHE. Clinicians believing
this to be the case might perceive that this review “misses
the point” of the PHE, by parsing out important parts of
the evaluation itself. While this is a valid and important
consideration, the heterogeneity in studies we reviewed re-
flects considerable disagreement within the medical com-
munity about what elements should be considered essential
to the PHE. In addition, a majority of studies identified
receipt of preventive health services as an outcome of the
PHE and not a part of the PHE. Furthermore, our defini-
tion of the PHE facilitated a comprehensive identification
of studies assessing its value and provided for a broad as-
sessment of many potential benefits and harms resulting
from the PHE. It is possible, however, that heterogeneity
in the PHE among studies could limit inferences regarding
which aspects of the PHE are most influential on some
outcomes.
Second, some of the largest trials assessing the PHE
were performed among select populations before publica-
tion of USPSTF guidelines in 1989, which may limit their
generalizability to current clinical practice. Heterogeneity
among studies’ reported outcomes prevented a systematic
analysis of any potential time trends.
Third, the feasibility of isolating the effect of the PHE
on long-term outcomes is unclear given the periodic (or
one-time) delivery of the PHE in studies and given multi-
ple other episodes of patient care that typically occur out-
side of the PHE. In one study, follow-up was as long as 20
years, with minimal analysis of potential competing causes
of long-term outcomes, limiting the ability to identify a
durable effect of the PHE (59).
Notwithstanding these limitations, our review pro-
vides a systematic appraisal of the literature identifying the
value of the PHE to date, which could serve as an impor-
tant foundation for clinical practice policies and future re-
search in this area.
In summary, this systematic review demonstrated that
the PHE has a beneficial effect on the delivery of some
clinical preventive services and may have a beneficial effect
on patient worry, providing justification for its continued
implementation in clinical practice. Further research is
needed to clarify the long-term benefits, harms, and costs
of undergoing the PHE and to weigh the value of receiving
clinical preventive services and worry relief in the absence
of evidence demonstrating such long-term clinical benefits.
Notwithstanding this, current evidence demonstrating
clear benefits of the PHE despite heterogeneity in the lit-
erature could provide clinicians with confidence that the
PHE may confer similar benefits in their own practices
while they use individual judgment regarding the optimal
way to implement the PHE.
From the Welch Center for Prevention, Epidemiology and Clinical
Research, Johns Hopkins School of Medicine, and Johns Hopkins
Bloomberg School of Public Health, Baltimore, Maryland.
Grant Support: By the Agency for Healthcare Research and Quality
contract 290-02-0018.
Potential Financial Conflicts of Interest: None disclosed.
Requests for Single Reprints: L. Ebony Boulware, MD, MPH, 2024
East Monument Street, Suite 2-600, Baltimore, MD 21205; e-mail,
lboulwa@jhmi.edu.
Current author addresses are available at www.annals.org.
References
1. Periodic health examination: a guide for designing individualized preventive
health care in the asymptomatic patients. Medical Practice Committee, American
College of Physicians. Ann Intern Med. 1981;95:729-32. [PMID: 7305155]
2. Merenstein D, Daumit GL, Powe NR. Use and costs of nonrecommended
tests during routine preventive health exams. Am J Prev Med. 2006;30:521-7.
[PMID: 16704947]
3. South-East London Screening Study Group. A controlled trial of multiphasic
screening in middle-age: results of the South-East London Screening Study.
1977. Int J Epidemiol. 2001;30:935-40. [PMID: 11689496]
4. Friedman GD, Collen MF, Fireman BH. Multiphasic Health Checkup Eval-
uation: a 16-year follow-up. J Chronic Dis. 1986;39:453-63. [PMID: 3711252]
5. Stone DH, D’Souza MF. Multiphasic screening in middle age: results and
Review Periodic Health Evaluation
298 20 February 2007 Annals of Internal Medicine Volume 146 • Number 4 www.annals.org
implications of a controlled trial in British general practice. Isr J Med Sci. 1981;
17:215-21. [PMID: 7228648]
6. Cutler JL, Ramcharan S, Feldman R, Siegelaub AB, Campbell B, Friedman
GD, et al. Multiphasic checkup evaluation study. 1. Methods and population.
Prev Med. 1973;2:197-206. [PMID: 4723635]
7. Patrick DL, Grembowski D, Durham M, Beresford SA, Diehr P, Ehreth J,
et al. Cost and outcomes of Medicare reimbursement for HMO preventive ser-
vices. Health Care Financ Rev. 1999;20:25-43. [PMID: 11482123]
8. Elder JP, Williams SJ, Drew JA, Wright BL, Boulan TE. Longitudinal effects
of preventive services on health behaviors among an elderly cohort. Am J Prev
Med. 1995;11:354-9. [PMID: 8775655]
9. Morrissey JP, Harris RP, Kincade-Norburn J, McLaughlin C, Garrett JM,
Jackman AM, et al. Medicare reimbursement for preventive care. Changes in
performance of services, quality of life, and health care costs. Med Care. 1995;
33:315-31. [PMID: 7731275]
10. Burton LC, Steinwachs DM, German PS, Shapiro S, Brant LJ, Richards
TM, et al. Preventive services for the elderly: would coverage affect utilization and
costs under Medicare? Am J Public Health. 1995;85:387-91. [PMID: 7892924]
11. Effectiveness of health checks conducted by nurses in primary care: final
results of the OXCHECK study. Imperial Cancer Research Fund OXCHECK
Study Group. BMJ. 1995;310:1099-104. [PMID: 7742676]
12. Faulkner LA, Schauffler HH. The effect of health insurance coverage on the
appropriate use of recommended clinical preventive services. Am J Prev Med.
1997;13:453-8. [PMID: 9415792]
13. U.S. Department of Health and Human Services. Prevention Makes Com-
mon “Cents.” Accessed at http://aspe.hhs.gov/health/prevention/ on 22 Septem-
ber 2006.
14. Chalmers TC, Smith H Jr, Blackburn B, Silverman B, Schroeder B, Reit-
man D, et al. A method for assessing the quality of a randomized control trial.
Control Clin Trials. 1981;2:31-49. [PMID: 7261638]
15. GRADE Working Group. Grading quality of evidence and strength of rec-
ommendations. BMJ. 2004;328:1490. [PMID: 15205295]
16. Boulware LE, Barnes GJ, Wilson RF, Phillips K, Maynor K, Hwang C, et
al. Value of the periodic health evaluation (Evidence Report/Technology Assess-
ment No. 136, prepared by Johns Hopkins Evidence-based Practice Center un-
der contract no. 290-02-0018). Rockville, MD: Agency for Healthcare Research
and Quality, U.S. Department of Health and Human Services; 2006. AHRQ
publication no. 06-E011.
17. Cohen J. Statistical power analysis for the behavioral sciences. 2nd ed. Hills-
dale, NJ: L. Erlbaum Associates; 1988.
18. Fletcher SW, Sourkes M, Rabzel M, Fletcher RH. Multiphasic screening.
Case-finding tool in a Teaching Hospital Medical Clinic. JAMA. 1977;237:887-
91. [PMID: 576328]
19. Belcher DW. Implementing preventive services. Success and failure in an
outpatient trial. Arch Intern Med. 1990;150:2533-41. [PMID: 2244769]
20. Nakanishi N, Tatara K, Fujiwara H. Do preventive health services reduce
eventual demand for medical care? Soc Sci Med. 1996;43:999-1005. [PMID:
8888469]
21. Theobald H, Bygren LO, Carstensen J, Hauffman M, Engfeldt P. Effects of
an assessment of needs for medical and social services on long-term mortality:
a randomized controlled study. Int J Epidemiol. 1998;27:194-8. [PMID:
9602398]
22. Grimaldi JV. The worth of occupational health programs: a new evaluation
of periodic physical examinations. J Gnathol. 1965;56:365-74. [PMID:
14321467]
23. Roberts NJ, Ipsen J, Elsom KO, Clark TW, Yanagawa H. Mortality among
males in periodic-health-examination programs. N Engl J Med. 1969;281:20-4.
[PMID: 5785733]
24. Slesinger DP, Tessler RC, Mechanic D. The effects of social characteristics
on the utilization of preventive medical services in contrasting health care pro-
grams. Med Care. 1976;14:392-404. [PMID: 1271880]
25. Geiger WJ, Neuberger MJ, Bell GC. Implementing the US preventive ser-
vices guidelines in a family practice residency. Fam Med. 1993;25:447-51.
[PMID: 8375602]
26. Kottke TE, Solberg LI, Brekke ML, Cabrera A, Marquez MA. Delivery
rates for preventive services in 44 midwestern clinics. Mayo Clin Proc. 1997;72:
515-23. [PMID: 9179135]
27. Sox CH, Dietrich AJ, Tosteson TD, Winchell CW, Labaree CE. Periodic
health examinations and the provision of cancer prevention services. Arch Fam
Med. 1997;6:223-30. [PMID: 9161346]
28. Bernacki EJ, Tsai SP, Malone RD. Participation in a periodic physical
examination program and group health care utilization and costs. J Occup Med.
1988;30:949-52. [PMID: 3230446]
29. Williams RB, Boles M, Johnson RE. A patient-initiated system for preven-
tive health care. A randomized trial in community-based primary care practices.
Arch Fam Med. 1998;7:338-45. [PMID: 9682687]
30. Hahn DL. The delivery of clinical preventive services: acute care intervention.
J Fam Pract. 1999;48:785-9. [PMID: 12224676]
31. Freedman A, Pimlott N, Naglie G. Preventive care for the elderly. Do family
physicians comply with recommendations of the Canadian Task Force on Pre-
ventive Health Care? Can Fam Physician. 2000;46:350-7. [PMID: 10690491]
32. Stange KC, Flocke SA, Goodwin MA, Kelly RB, Zyzanski SJ. Direct ob-
servation of rates of preventive service delivery in community family practice. Prev
Med. 2000;31:167-76. [PMID: 10938218]
33. Hama Y, Masumori K, Tagami H, Fujiwara K, Kusano S. Preassignment
examination for personnel on Iwo Jima. Mil Med. 2001;166:721-4. [PMID:
11515325]
34. Nutting PA, Baier M, Werner JJ, Cutter G, Conry C, Stewart L. Compet-
ing demands in the office visit: what influences mammography recommenda-
tions? J Am Board Fam Pract. 2001;14:352-61. [PMID: 11572540]
35. Parchman M, Byrd T. Access to and use of ambulatory health care by a
vulnerable Mexican American population on the U.S.-Mexico border. J Health
Care Poor Underserved. 2001;12:404-14. [PMID: 11688192]
36. Tao G, Zhang P, Li Q. Services provided to nonpregnant women during
general medical and gynecologic examinations in the United States. Am J Prev
Med. 2001;21:291-7. [PMID: 11701300]
37. Burton WN, Chen CY, Conti DJ, Schultz AB, Edington DW. The value of
the periodic executive health examination: experience at Bank One and summary
of the literature. J Occup Environ Med. 2002;44:737-44. [PMID: 12185794]
38. Chiou CJ, Chang HY. Do the elderly benefit from annual physical exami-
nation? An example from Kaohsiung City, Taiwan. Prev Med. 2002;35:264-70.
[PMID: 12202069]
39. Finkelstein MM. Preventive screening. What factors influence testing? Can
Fam Physician. 2002;48:1494-501. [PMID: 12371308]
40. Schneider GW, DeHaven M, Snell LM. Fostering a culture of prevention in
a residency program through a continuous quality improvement project. Am J
Med Qual. 2003;18:82-9. [PMID: 12710557]
41. Flocke SA, Stange KC. Direct observation and patient recall of health behav-
ior advice. Prev Med. 2004;38:343-9. [PMID: 14766118]
42. Lin SX, Gebbie KM, Fullilove RE, Arons RR. Do nurse practitioners make
a difference in provision of health counseling in hospital outpatient departments?
J Am Acad Nurse Pract. 2004;16:462-6. [PMID: 15543924]
43. Somkin CP, McPhee SJ, Nguyen T, Stewart S, Shema SJ, Nguyen B, et al.
The effect of access and satisfaction on regular mammogram and Papanicolaou
test screening in a multiethnic population. Med Care. 2004;42:914-26. [PMID:
15319618]
44. Burton LC, German PS, Shapiro S. A preventive services demonstration.
Health status, health behaviors, and cost outcomes 2 years after intervention. The
Johns Hopkins Medicare Preventive Services Demonstration Team. Med Care.
1997;35:1149-57. [PMID: 9366893]
45. Burton LC, Paglia MJ, German PS, Shapiro S, Damiano AM. The effect
among older persons of a general preventive visit on three health behaviors: smok-
ing, excessive alcohol drinking, and sedentary lifestyle. The Medicare Preventive
Services Research Team. Prev Med. 1995;24:492-7. [PMID: 8524724]
46. Collen MF, Dales LG, Friedman GD, Flagle CD, Feldman R, Siegelaub
AB. Multiphasic checkup evaluation study. 4. Preliminary cost benefit analysis for
middle-aged men. Prev Med. 1973;2:236-46. [PMID: 4723639]
47. Patrick DL, Grembowski D, Durham M, Beresford SA, Diehr P, Ehreth J,
et al. Cost and outcomes of Medicare reimbursement for HMO preventive ser-
vices. Health Care Financ Rev. 1999;20:25-43. [PMID: 11482123]
48. Dales LG, Friedman GD, Collen MF. Evaluating periodic multiphasic
health checkups: a controlled trial. J Chronic Dis. 1979;32:385-404. [PMID:
109452]
49. Sheridan S, Pignone M, Donahue K. Screening for high blood pressure: a
review of the evidence for the U.S. Preventive Services Task Force. Am J Prev
Med. 2003;25:151-8. [PMID: 12880884]
50. Dijkstra A, Brosschot J. Worry about health in smoking behaviour change.
Behav Res Ther. 2003;41:1081-92. [PMID: 12914809]
51. Hvas L, Reventlow S, Jensen HL, Malterud K. Awareness of risk of osteo-
porosis may cause uncertainty and worry in menopausal women. Scand J Public
ReviewPeriodic Health Evaluation
www.annals.org 20 February 2007 Annals of Internal Medicine Volume 146 • Number 4 299
Health. 2005;33:203-7. [PMID: 16040461]
52. Wilcox S, Ainsworth BE, LaMonte MJ, DuBose KD. Worry regarding
major diseases among older African-American, Native-American, and Caucasian
women. Women Health. 2002;36:83-99. [PMID: 12539794]
53. McCaul KD, Schroeder DM, Reid PA. Breast cancer worry and screening:
some prospective data. Health Psychol. 1996;15:430-3. [PMID: 8973922]
54. Harris R, Donahue K, Rathore SS, Frame P, Woolf SH, Lohr KN. Screen-
ing adults for type 2 diabetes: a review of the evidence for the U.S. Preventive
Services Task Force. Ann Intern Med. 2003;138:215-29. [PMID: 12558362]
55. Pignone M, Rich M, Teutsch SM, Berg AO, Lohr KN. Screening for
colorectal cancer in adults at average risk: a summary of the evidence for the U.S.
Preventive Services Task Force. Ann Intern Med. 2002;137:132-41. [PMID:
12118972]
56. Humphrey LL, Helfand M, Chan BK, Woolf SH. Breast cancer screening:
a summary of the evidence for the U.S. Preventive Services Task Force. Ann
Intern Med. 2002;137:347-60. [PMID: 12204020]
57. Yarnall KS, Pollak KI, Østbye T, Krause KM, Michener JL. Primary care:
is there enough time for prevention? Am J Public Health. 2003;93:635-41.
[PMID: 12660210]
58. McCarthy BD, Yood MU, MacWilliam CH, Lee MJ. Screening mammog-
raphy use: the importance of a population perspective. Am J Prev Med. 1996;12:
91-5. [PMID: 8777073]
59. A controlled trial of multiphasic screening in middle-age: results of the South-
East London Screening Study. The South-East London Screening Study Group.
Int J Epidemiol. 1977;6:357-63. [PMID: 608798]
Review Periodic Health Evaluation
300 20 February 2007 Annals of Internal Medicine Volume 146 • Number 4 www.annals.org
Current Author Addresses: Dr. Boulware: 2024 East Monument Street,
Suite 2-600, Baltimore, MD 21205.
Dr. Marinopoulos: 10753 Falls Road, Suite 325, Lutherville, MD
21093.
Dr. Phillips: 1830 East Monument Street, Room 8033, Baltimore, MD
21205.
Dr. Hwang: 10490 Little Patuxent Parkway, Suite 610, Columbia, MD
21044.
Dr. Maynor: Geisinger Medical Center, 1000 East Mountain Boulevard,
Wilkes-Barre, PA 18711.
Dr. Merenstein: 215 Kober Cogan Hall, 3750 Reservoir Road NW,
Washington, DC 20007.
Ms. Wilson, MSc 1830 East Monument Street, Room 8061, Baltimore,
MD 21287.
Mr. Barnes: 525 West Redwood Street, Baltimore, MD 21201.
Dr. Bass: 1830 East Monument Street, Room 8068, Baltimore, MD
21287.
Dr. Powe: 2024 East Monument Street, Suite 2-600, Baltimore, MD
21205.
Dr. Daumit: 2024 East Monument Street, Suite 2-500, Baltimore, MD
21205.
Annals of Internal Medicine
W-74 20 February 2007 Annals of Internal Medicine Volume 146 • Number 4 www.annals.org
Appendix Table. Effect Sizes in Randomized, Controlled Trials by Outcome Assessed*
Outcomes Effect Size (95% CI)† (Reference) Reference for
Which Effect
Size Could
Not Be
Calculated§
In Studies with Positive Effect
of PHE‡
In Studies with Negative Effect of PHE Confidence Interval
Crosses 0
Receipt of Papanicolaou smear 1.71 (1.69 to 1.73) (9)
0.07 (0.07 to 0.07) (44)
Preventive counseling 1.09 (1.08 to 1.11) (19)
a
1.19 (1.17 to 1.21) (19)
b
Immunizations 0.35 (0.33 to 0.36) (9)
c
0.10 (0.10 to 0.10) (7)
d
–0.22 (–0.24 to –0.20) (19)
e
Cholesterol screening 0.02 (0.00 to 0.04) (9)
Colon cancer screening
(fecal occult blood testing)
1.19 (1.17 to 1.21) (9)
1.07 (1.05 to 1.08) (19)
Mammography 0.14 (0.12 to 0.16) (9)
Disease detection 0.03 (0.02 to 0.03) (3)
f
0.96 (0.84 to 1.08) (18)
g
0.53 (0.41 to 0.64) (18)
h
–0.01 (–0.01 to –0.01) (3)
i
–0.03 (–0.03 to –0.03) (3)
j
–0.01 (–0.01 to 0.00) (3)
k
Health habits 0.28 (0.14 to 0.42) (8)
l
0.120 (0.117 to 0.123) (7)
m
0.040 (0.037 to 0.043) (7)
n
0.345 (0.342 to 0.348) (7)
o
0.080 (0.077 to 0.083) (7)
p
0.020 (0.017 to 0.023) (7)
q
0.020 (0.017 to 0.023) (7)
r
0.100 (0.098 to 0.102) (11)
s
0.032 (0.030 to 0.034) (11)
t
0.088 (0.086 to 0.090) (11)
u
0.244 (0.242 to 0.246) (11)
v
0.250 (0.248 to 0.252) (11)
w
0.13 (0.11 to 0.14) (45)
x
–0.040 (–0.043 to –0.037) (7)
y
–0.014 (–0.016 to –0.012) (3)
z
–.02 (–.03 to –.02) (45)
aa
0.000 (–0.14 to 0.14) (8)
bb
0.01 (–0.13 to 0.15) (8)
cc
0.02 (–0.12 to 0.16) (8)
dd
0.05 (–0.09 to 0.19) (8)
ee
0.01 (–0.13 to 0.15) (8)
ff
Patient attitudes 7
Health status 10
Blood pressure 0.12 (0.02 to 0.21) (8)
gg
0.11 (0.04 to 0.18) (11)
hh
0.13 (0.06 to 0.19) (11)
ii
0.022 (0.019 to 0.024) (11)
jj
0.03 (–0.06 to 0.13) (8)
kk
Changes in serum cholesterol
levels
0.22 (0.16 to 0.29) (11)
ll
0.09 (0.09 to 0.10) (11)
mm
Body mass index 0.087 (0.022 to 0.153) (11)
nn
0.032 (0.030 to 0.034) (11)
oo
–0.020 (–0.023 to –0.017) (7)
pp
–0.031 (– 0.170 to 0.108) (8)
qq
–0.036 (–0.174 to 0.103) (8)
rr
Reduction in health care costs 0.06 (–0.03 to 0.15) (9)
ss
0.05 (–0.04 to 0.14) (9)
tt
7, 10, 46
Reduction in disability 0.060 (0.054 to 0.066) (48)
uu
–0.014 (–0.016 to –0.012) (3)
Reduction in hospitalizations 0.01 (0.00 to 0.01) (3)
vv
0.02 (–0.07 to 0.11) (9)
ww
–0.04 (–0.13 to 0.05) (9)
xx
10
yy,zz,aaa,bbb
Reduction in all-cause
mortality
0.06 (0.05 to 0.06) (10)
ccc
0.004 (0.004 to 0.005) (4)
ddd
–0.03 (–0.04 to –0.03) (7)
eee
–0.002 (–0.003 to –0.0003) (3)
fff
Rate ratio:
1.03 (0.94 to 1.14) (21)
* PHE ⫽ periodic health evaluation.
† Magnitude and direction of effect of receipt of PHE on outcome, based on standardized effect sizes calculated by using the Cohen d statistic. We considered effect sizes
ranging from 0 to 0.25 to represent small effects, those ranging from 0.26 to 0.8 to represent intermediate effects, and those greater than 0.8 to represent large effects. Effect
sizes can be thought of as the average percentile standing of the average treated (or experimental) participant relative to the average untreated (or control) participant. An effect
size of 0.0 indicates that the mean of the treated group is at the 50th percentile of the untreated group. An effect size of 0.25 indicates that the mean of the treated group
is at the 58th percentile of the untreated group. An effect size of 0.8 indicates that the mean of the treated group is at the 79th percentile of the untreated group.
‡ The same studies reported on multiple outcomes. Where a letter superscript is indicated next to a citation, additional information regarding the reported outcome is listed
here: a: smoking cessation; b: alcohol abuse; c: influenza vaccination; d: influenza vaccination; e: influenza vaccination; f: ischemia on electrocardiogram; g: detection of “all
problems” before and after intervention; h: disease detection of “important problems” before and after intervention; i: angina; j: bronchitis symptoms; k: high diastolic blood
pressure; l: fiber servings per day; m: physical activity; n: diet (fat and fiber); o: advance directives; p: breast self-examination; q: smoking; r: alcohol use; s: smoking; t: alcohol
use; u: exercise less than once per month; v: use full-cream milk; w: use butter or hard margarine; x: smoking; y: seat belt use; z: percentage still smoking; aa: problem alcohol
drinking; bb: fat servings per week; cc: salt use; dd: caffeine drinks per day; ee: stretching minutes per week; ff: consumption of cruciferous foods; gg: mean systolic blood
pressure at 12 months’ follow-up; hh: systolic blood pressure at 3 years’ follow-up; ii: diastolic blood pressure at 3 years’ follow-up; jj: proportion of high-risk diastolic pressure
(ⱖ100 mm Hg) at 3 years’ follow-up; kk: mean diastolic blood pressure at 12 months’ follow-up; ll: mean total cholesterol at 3 years’ follow-up; mm: proportion with “high
risk” cholesterol level (ⱖ8 mmol/L) at 3 years’ follow-up; nn: mean body mass at 3 years’ follow-up; oo: proportion of participants with body mass index ⱖ30 kg/m
2
; pp:
at risk for obesity at 24 months’ follow-up; qq: mean body mass index at 24 months’ follow-up; rr: mean body mass index at 48 months’ follow-up; ss: 3-year postintervention
cumulative Medicare charges; tt: 3-year postintervention cumulative Medicare reimbursement; uu: disability at 11 years’ follow-up; vv: hospitalizations; ww: hospital days per
enrollee; xx: admissions per enrollee; yy: mean inpatient days for the intervention and control groups who had a hospital discharge in that year (year 1); zz: mean inpatient
days (year 2); aaa: hospital discharges per 1000 (year 1); bbb: hospital discharges per 1000 (year 2); ccc: death; ddd: deaths, rate per 1000 persons at 16 years; eee: mortality
at 48 months’ follow-up; fff: mortality rate per 1000 person-years at risk.
§ Standardized effect size could not be calculated for the study or some studies assessing this outcome.
www.annals.org 20 February 2007 Annals of Internal Medicine Volume 146 • Number 4 W-75