Client-Directed Interventions to Increase Community Demand for

CDCARJSS

A

I

Cd

FMaCWCC(P

t

G3

GN

rPaN

S

lient-Directed Interventions to Increase Communityemand for Breast, Cervical, and Colorectalancer ScreeningSystematic Review

oy C. Baron, MD, MPH, Barbara K. Rimer, DrPH, Rosalind A. Breslow, PhD, Ralph J. Coates, PhD,on Kerner, PhD, Stephanie Melillo, MPH, Nancy Habarta, MPH, Geetika P. Kalra, MPH,ajal Chattopadhyay, PhD, MPH, Katherine M. Wilson, PhD, Nancy C. Lee, MD, Patricia Dolan Mullen, DrPH,teven S. Coughlin, PhD, MPH, Peter A. Briss, MD, and the Task Force on Community Preventive Services

bstract: Most major medical organizations recommend routine screening for breast, cervical, and colorectalcancers. Screening can lead to early detection of these cancers, resulting in reduced mortality. Yetnot all people who should be screened are screened, either regularly or, in some cases, ever. Thisreport presents the results of systematic reviews of effectiveness, applicability, economic efficiency,barriers to implementation, and other harms or benefits of interventions designed to increasescreening for breast, cervical, and colorectal cancers by increasing community demand for theseservices. Evidence from these reviews indicates that screening for breast cancer (mammography)and cervical cancer (Pap test) has been effectively increased by use of client reminders, small media,and one-on-one education. Screening for colorectal cancer by fecal occult blood test has beenincreased effectively by use of client reminders and small media. Additional research is needed todetermine whether client incentives, group education, and mass media are effective in increasinguse of any of the three screening tests; whether one-on-one education increases screening forcolorectal cancer; and whether any demand-enhancing interventions are effective in increasing theuse of other colorectal cancer screening procedures (i.e., flexible sigmoidoscopy, colonoscopy,double contrast barium enema). Specific areas for further research are also suggested in this report.(Am J Prev Med 2008;35(1S):S34–S55) © 2008 American Journal of Preventive Medicine

awftBdcAvUypraioLapafiw

ntroduction

ancer is a major public health problem in theU.S. In 2003, more than 1,290,000 people werediagnosed with cancer and more than 556,000

ied of cancer.1,a This included more than 55,000 men

rom the Community Guide Branch, National Center for Healtharketing (Baron, Melillo, Habarta, Kalra, Chattopadhyay, Briss)

nd Division of Cancer Prevention and Control, National Center forhronic Disease Prevention and Health Promotion (Breslow, Coates,ilson, Lee, Coughlin), CDC, Atlanta, Georgia; University of Northarolina School of Public Health (Rimer), Chapel Hill, Northarolina; National Cancer Institute, National Institutes of HealthKerner), Bethesda, Maryland; and University of Texas School ofublic Health (Mullen), Houston, Texas.Author affiliations are shown at the time the research was conducted.The names and affiliations of the Task Force members are listed at

he front of this supplement and at www.thecommunityguide.org.Address correspondence to Roy C. Baron, MD, MPH, Community

uide Branch, CDC, 1600 Clifton Road NE, MS E-69, Atlanta GA0333. E-mail: [email protected] reprint requests to Shawna L. Mercer, MSc, PhD, The

uide to Community Preventive Services, CDC, 1600 Clifton RoadE, MS E-69, Atlanta GA 30333. E-mail: [email protected] of cancer diagnoses are based on the most current

eports of observed cases from cancer registries in CDC’s Nationalrogram of Cancer Registries and NCI’s Surveillance, Epidemiology,

gnd End Results Program. Numbers of deaths are from CDC’sational Vital Statistics Program.

34 Am J Prev Med 2008;35(1S)© 2008 American Journal of Preventive Medicine • Publish

nd women who died from colorectal cancer, 41,000omen from breast cancer, and nearly 4000 women

rom cervical cancer. According to a 2003 report fromhe Institute of Medicine’s National Cancer Policyoard,2 each year 4475 deaths from breast cancer, 3644eaths from cervical cancer, and 9632 deaths fromolorectal cancer could be prevented if all eligiblemericans received appropriate cancer screening ser-ices. Yet the 2005 National Health Interview Survey of.S. adults3 found that only 67% of women aged �40

ears reported having had mammograms within therevious 2 years, and 78% of women aged �18 yearseported Pap tests within the previous 3 years. Amongdults aged �50, only 50% reported ever having screen-ng endoscopies and only 17% reported having fecalccult blood tests (FOBT) within the previous 2 years.ower rates were observed among American Indiansnd Alaska Natives; people of Asian, Latino, or His-anic ethnicity; African Americans (endoscopy, only);nd among poor and less-educated populations. Ratesor recommended screenings tend to be lower amongndividuals without a usual source of health care,ithout health insurance, and among recent immi-
rants to the U.S.4 At the same time, efforts to maxi-
0749-3797/08/$–see front mattered by Elsevier Inc. doi:10.1016/j.amepre.2008.04.002

http://www.thecommunityguide.org

mailto:[email protected]

mailto:[email protected]

mtaptrs

tpcdhhtic

GFhamiaaogidttattdEFsn

ntmr(cdaatmr

itc

rnl

M

Grfcwotsssotsteetsqa

sttsopifldcaistaaiTaactsu

amet

aw

J

ize control of breast, cervical, and colorectal cancershrough screening face the additional challenge ofssuring that cancer screening, once initiated, is re-eated at recommended intervals.5,6 Increasing use ofhese screening tests at recommended intervals andeducing inequalities in screening use are importantteps toward reducing cancer morbidity and mortality.2

An array of community- and systems-based interven-ions are available to programs and planners for use inromoting cancer screening.7,8 These interventionsan target clients (client-directed), providers (provider-irected), or both, each either directly or through theealthcare system. Many of these interventions alsoave been applied in other areas of public health, but

heir effectiveness, applicability, and cost effectivenessn increasing cancer screening rates are either notlearly established or not completely understood.

The Guide to Community Preventive Services (Communityuide), developed by the independent, nonfederal Taskorce on Community Preventive Services (Task Force),as conducted systematic reviews on the effectiveness,pplicability, economic efficiency, barriers to imple-entation, and other harms or benefits of community

nterventions to increase screening for breast, cervical,nd colorectal cancers.7 The conceptual approach tond selection of interventions for these reviews focusedn three primary strategies to close screening-relatedaps: increasing community demand for cancer screen-ng services, reducing barriers to access, and increasingelivery of these services by healthcare providers. The firstwo strategies encompass client-directed approaches in-ended to influence client knowledge, motivation, access,nd decision to be screened at appropriate intervals. Thehird strategy encompasses provider-directed approacheso reduce missed opportunities to recommend, order, oreliver cancer screening services at appropriate intervals.vidence from these reviews provides the basis for Taskorce recommendation of interventions in each of thesetrategic areas as well as for identifying additional researcheeds.In this report, evidence is reviewed on the effective-

ess of classes of client-directed interventions intendedo increase community demand for screening recom-

ended for early detection of breast cancer (mammog-aphy), cervical cancer (Pap test), and colorectal cancerguaiac-based FOBT, flexible sigmoidoscopy, colonos-opy, or double contrast barium enema).9–12 Client-irected interventions designed to increase communityccess to these services are reviewed in an accompanyingrticle,13 as are two types of provider-directed interven-ions.14 An additional provider-directed intervention and

ulticomponent (combinations of) interventions will beeviewed in future publications.

The use of community will usually refer to a group ofndividuals who share one or more characteristics,15 inhis case the potential to benefit from one or more
ancer screening services. Community is also used in t
uly 2008

eference to a setting or in combination with “commu-ity healthcare worker,” in which case the intent is

ocale, neighborhood, or other geopolitical unit.

ethods

eneral methods for conducting Community Guide systematiceviews have been described in detail.16,17 Specific methodsor conducting reviews of interventions to increase breast,ervical, and colorectal cancer screening are described else-here in this supplement.8 That description includes theverall literature search of primary scientific publicationshrough November 2004, selection of the 244 candidatetudies satisfying general inclusion criteria for the cancercreening reviews, and specific criteria applied to the finalelection of qualifying studies for each review (suitabilityf study design and quality of execution16; see Results sec-ions). In this section, methodologic issues are briefly discussed,pecific to classes of interventions covered in this article, that is,hose designed to increase community demand—client remind-rs or recall, client incentives, small media, mass media, groupducation, and one-on-one education—and for which 128 ofhe 244 candidate studies were considered for review. Aummary of the results and other details of the finalualifying studies for each intervention review are availablet www.thecommunityguide.org/cancer.The analytic model (Figure 1), similar to other con-

tructs used in Community Guide cancer screening interven-ion reviews,7,8,13,14 shows hypothesized relationships be-ween interventions to increase community demand, aeries of intermediate steps, and ultimate (desired) healthutcomes. Completed screening (shaded) is the outcome ofrimary interest in these reviews. Although completed screen-

ng is an intermediate step in the model, it provides the basisor evaluation of intervention effectiveness because of estab-ished links to the health outcome of ultimate interest:ecreased mortality from breast, cervical, and colorectalancers.9–12 Interventions to increase community demandre directed toward age-eligible populations with the goal ofncreasing adherence to screening recommendations. Theystematic review development team (the team)7 postulatedhat by positively influencing some combination of knowledge,wareness, and intent (the last of which may require alteringttitudes and beliefs about screening services and tests), eachntervention has the potential to increase demand for screening.hese changes, in turn, would lead to increased test completionnd early detection and, ultimately, reduce cancer morbiditynd mortality. The interventions might also cue or promptlients who are ready for screening. The model also indicateshat these interventions may result in other benefits and harms,uch as positive or negative effects on other health behaviors orse of healthcare services.Although several recommended screening procedures are

lso used for diagnostic or therapeutic purposes (i.e., mam-ography, colorectal endoscopy, and double contrast barium

nema), reference to them in these reviews relates specificallyo the screening application.

Intervention effectiveness was evaluated by comparing pre-nd post-intervention screening practices in the study groupsith those in groups receiving no intervention. For each study,

he measure of effect was represented, where possible, as per-

Am J Prev Med 2008;35(1S) S35

http://www.thecommunityguide.org/cancer

cfacgrp

ivqihCpohsatcd

eTcrei

R

Cc(psmftf

sTiocarp

B

Escmcwdtqe

otTTsetvt

atimes

Fio

S

entage point (i.e., absolute) change in completed screeningrom baseline or comparison value. When an effect was reporteds an odds ratio (OR) or percent (i.e., relative) change andould not be converted to percentage point change, it wasenerally excluded from the summary effect measure. Theseesults were reported separately, however, to reflect the com-lete evidence base and to assess consistency across all studies.In general, to answer questions about whether particular

nterventions are effective, Community Guide systematic re-iews consider data from all available studies of sufficientuality that compare outcomes in a group exposed to an

ntervention with outcomes in a group either concurrently oristorically unexposed (or less exposed) to the intervention.16,17

onsistent with many groups that focus on population-based orublic health interventions,18 this approach is broadly inclusivef a range of study designs. (For the review of client reminders,owever, a large number of studies had designs of greatestuitability [i.e., those with pre- and post-intervention assessmentsnd comparison groups]. The review of that intervention washerefore limited to studies of greatest design suitability; ouronclusion was that exclusion of other levels of design suitabilityid not compromise external validity.)As noted elsewhere in this supplement,7,8 client-related barri-

rs can differ by screening test and by population subgroup.herefore, effectiveness, applicability, and economic effi-iency of client-directed interventions were reviewed sepa-ately for breast, cervical, and colorectal cancers. Also consid-red were other positive or negative effects, barriers tomplementation, and areas needing further research.

esults: Client Reminders

lient reminder or recall (referred to collectively aslient reminders) included in this review are printedletter or postcard) or telephone messages advisingeople that they are due (reminder) or late (recall) forcreening. Client reminders, as defined by our team,ay be enhanced by one or more of the following: a

ollow-up printed or telephone reminder; additionalext or discussion with information about indications

igure 1. Analytic framework: interventions to increase comnterventions; rectangles with rounded corners indicate medutcomes.)

or, benefits of, and ways to overcome barriers to e

36 American Journal of Preventive Medicine, Volume 35, Num

creening; or assistance in scheduling appointments.ailored reminders (printed or verbal) address the

ndividual’s risk profile or other relevant psychologicalr social characteristics, such as what keeps a specificlient from seeking screening and what would encour-ge the client to be screened. The effectiveness of clienteminders for improving adherence to several otherreventive interventions is well documented.19–23

reast Cancer

ffectiveness. Thirty-nine studies24–62 of greatest de-ign suitability were identified that reported usinglient reminders to increase breast cancer screening byammography. Of these, nine studies24–32 were ex-

luded due to limited quality of execution8 and 1133–43

ere excluded because comparison groups receivedifferent reminders or reminders of lesser intensityhan study groups. Of the 19 remaining studies thatualified for review, 1744–50,52–60,62 had fair quality ofxecution and two51,61 had good quality of execution.All studies enrolled eligible women who were due or

verdue for mammography and assessed screening comple-ion using self-reports57,60 or record reviews.44–56,58,59,61,62

he 19 qualifying studies evaluated 32 intervention arms.hree studies56,57,61 evaluated three interventions, four

tudies44,45,53,58 evaluated two interventions (one58 studyvaluated each intervention at two separate screening loca-ions), and 12 studies46–52,54,55,59,60,62 evaluated one inter-ention (one study46 evaluated the intervention at two dis-inct locations).

Printed reminders were either usedlone44–46,48,49,53–58,60,62 or enhanced by one or more ofhe following elements: face-to-face counseling, schedul-ng assistance or direct referral, follow-up telephone re-

inder (with or without scheduling assistance or anducational component), or a follow-up letter offeringcheduling assistance.45,50–53,57–59,62 Telephone remind-

ity demand for cancer screening services. (Oval indicatess or intermediate outcomes, and rectangle indicates health

muniator

rs were either used alone,56 with scheduling assistance

ber 1S www.ajpm-online.net

(pii

c(tstsia(rvIbrwn

AectacAms

remCst

w2ppeapistecptetwr

Egce

Fv one r

J

with47 or without44 education), or with a second tele-hone reminder offering scheduling assistance.61 Five

nterventions included tailored (barrier-specific) counsel-ng, four by telephone47,57,61 and one in person.57

Overall, the median post-intervention increase inompleted mammography was 14.0 percentage pointsinterquartile interval [IQI]�2.0, 24.0). The magni-ude of this effect and consistent positive results acrosstudies and reminder systems demonstrate the effec-iveness of client reminders in increasing breast cancercreening by mammography. Nonetheless, as indicatedn Figure 2, simple printed reminders, when usedlone, appear to result in effects of smaller magnitudemedian 3.6 percentage points; IQI�1.8, 14.0) thaneminders used with additional components or con-eyed by telephone (median 18.5 percentage points;QI�10.5, 32.0). This observation is further supportedy all nine intrastudy comparisons identified in thiseview,44,45,53,56–58 which are shown in Figure 2. Thereas no single component or combination of compo-ents found to account for larger effect sizes.

pplicability. The same body of evidence was used tovaluate the applicability of client reminders for in-reasing breast cancer screening in different popula-ions and settings. Where population and setting char-cteristics were specified, effective interventions wereonducted in the U.S. and Australia, among Africanmericans and whites, and in populations of low toixed or middle-class SES. Certain major population

igure 2. Percentage point change in completed mammogramersus printed reminders plus additional components or teleph

ubgroups, including Hispanics and Asians, had limited r

uly 2008

epresentation in these studies. Interventions were alsoffective in HMOs and other clinical settings, in com-unity settings, and in both rural and urban locations.lient reminders should be applicable across a range of

ettings and populations, provided they are adapted toarget populations and delivery context.

Only two studies reported on populations of womenho had never been screened. One59 demonstrated a4 percentage point increase in adherence followingrinted reminders delivered with prescheduled ap-ointments (reminders of lesser intensity were notvaluated). The other57 showed that printed reminderslone were effective when delivered to women with arior history of mammography (19.0 percentage point

ncrease, p�0.05) but not to those who had never beencreened (�3.4 percentage point change). At the sameime, the study demonstrated these reminders wereffective in both groups (23.7 percentage point in-rease, p�0.05, and 45.5 percentage point increase,�0.09, respectively) when they included an educa-ional component. Such findings suggest the utility ofnhancing simple reminders with additional interven-ion elements when targeting populations of womenho have never been screened or who may be hard toeach (see Research Issues, Client Reminders).

conomic efficiency. Six studies, five classified asood45,47,56,57,62 and one as very good51 met inclusionriteria8 for cost-effectiveness analysis of client remind-rs in increasing breast cancer screening by mammog-

ning attributable to client reminders (printed reminders onlyeminders). IQI, interquartile interval; NS, nonsignificant

scree

aphy. For one study,57 cost effectiveness reported in


tipvstt(asmatigmothEbatada(maps

C

Eou

ietriqe

hsttetvvmrtaatt

Ppnaebiwab

FN

S

erms of cost per percentage point increase in screen-ng rates was converted by Community Guide staff to coster additional screening examination. From ten inter-ention arms studied, estimated cost per additionalcreening ranged from $4.89 to $100.61. One study withhree intervention arms56 based estimates on a simpleelephone reminder delivered by nonphysician staff$4.89), a printed reminder from the physician ($43.12),nd a telephone reminder by the physician ($100.61). Aecond study with three intervention arms57 based esti-ates on a reminder letter from a physician ($7.66) andsimilar letter supplemented by face-to-face ($62.16) or

elephone ($71.85) counseling. Two separate studies us-ng similar methods and reported by the same investi-ators estimated cost effectiveness of letters signed byembers of group general practices and mailed with

ther educational materials to women being advisedhat screening was due ($47.01)51 and to women whoad not responded to an earlier notice ($63.28).62

ach of these estimates was likely inflated by includingoth management of increased clinic attendance anddditional consultation costs, which are beyond those ofhe intervention to promote screening. Finally, one studyssessed cost effectiveness of a letter from a radiologyepartment followed by a phone call ($9.56)45 and onessessed a telephone reminder with counseling$22.00).47 The range of estimates from these studiesade it difficult to ascertain the most cost-effective

pproach although, based on relatively few studies,hysician participation and enhancement of reminderseem to add to the cost of intervention.

ervical Cancer

ffectiveness. Twenty-one studies24–26,39,48,50,52,61,63–75

f greatest design suitability were identified that reported

igure 3. Percentage point change in completed Pap test screS, nonsignificant

sing client reminders to increase cervical cancer screen- c


ng by Pap test. Of these, eight studies24–26,63–67 werexcluded due to limited quality of execution andwo39,68 were excluded because comparison groupseceived different reminders or reminders of lesserntensity than study groups. Of 11 remaining studiesualifying for review, ten48,50,52,69–75 had fair quality ofxecution and one61 had good quality of execution.Most of the 11 qualifying studies enrolled women who

ad not been screened in at least 3 years, although one72

pecified 5 years, one73 specified at least 2 years, andwo71,75 specified 1 year. Pap test completion was ascer-ained in each study by record reviews. Qualifying studiesvaluated 15 intervention arms. One study61 evaluatedhree interventions, two studies69,71 evaluated two inter-entions, and the remaining studies evaluated one inter-ention. Nine intervention arms included printed re-inders alone48,50,52,69,70,72–75; three included printed

eminders combined with a follow-up printed61,71 orelephone61 reminder, with61 or without71 schedulingssistance; and three included telephone reminderslone69 or combined with either scheduling assis-ance alone61 or with scheduling assistance and aailored (barrier-specific) educational component.71

Overall, the median post-intervention increase inap test completion over 14 intervention arms was 10.2ercentage points (IQI�6.3, 17.9; Figure 3). The mag-itude of this effect and the consistency across studiesnd reminder systems demonstrate that client remind-rs are effective in increasing cervical cancer screeningy Pap test. Effectiveness is further supported by find-

ngs from an evaluation of a printed reminder alone,74

hich showed a statistically significant OR in a favor-ble direction but was not included in the analysisecause it could not be converted to a percentage point

g attributable to client reminders. IQI, interquartile interval;
enin
hange.


etcpo

AectuCAlitaaeatcstiwwpspotC

EsrPatfb(rwdoTldaebuica

iicTii

C

Edcisblbmgh

oteatvrrfaba

d1mrscfe

Aecesistgsgbro

J

A smaller effect was observed from printed remind-rs used alone (n�8, median 9.8 percentage points)han from reminders with additional components oronveyed by telephone (n�6, median 15.5 percentageoints). This difference, however, is supported by onlyne available intrastudy comparison.69

pplicability. The same body of evidence was used tovaluate the applicability of client reminders for cervi-al cancer screening in different populations and set-ings. Effective intervention studies that specified pop-lation and setting were conducted in the U.S.,anada, Australia, and the United Kingdom, amongfrican Americans and whites, and in populations of

ow SES. Certain major population subgroups, includ-ng Hispanics and Asians, had limited representation inhese studies. Interventions were also effective in HMOsnd a large medical practice, in community settings,nd in both urban and rural locations. Client remind-rs for cervical cancer screening should be applicablecross a range of settings and populations, providedhey are adapted to target populations and deliveryontext. Only one study72 reported an outcome for aubset of women with no previous Pap test history. Inhis study, a simple printed reminder was effective inncreasing the Pap test completion rate among theseomen (13 percentage points, p�0.05) and amongomen whose last Pap tests were �5 years earlier (22ercentage points, p�0.05). However, because fewtudies are available to assess these interventions inopulations of women who have never been screenedr who may be hard to reach, questions remain aboutheir applicability in these groups (see Research Issues,lient Reminders).

conomic efficiency. Four studies69,71,74,76 met inclu-ion criteria8 for cost-effectiveness analysis of clienteminders in increasing cervical cancer screening byap test. Three studies69,71,74 were classified as goodnd one76 was classified as very good. In seven interven-ion arms studied, cost per additional screening rangedrom $14.27 to $151.00. The lowest estimates wereased on simple printed ($14.27) and telephone$17.11) reminders in one study,69 and on a printedeminder that included an informational brochure andas later followed by a second reminder to nonrespon-ents ($18.69) and a telephone reminder that includedne-on-one counseling ($17.03) in another study.71

wo intermediate estimates were based on reminderetters that did ($85.23) or did not ($100.36) provide aesignated appointment time; however, these estimateslso included substantial fixed costs.76 The highest cost-ffectiveness estimate, $151.00,74 was from a population-ased personal printed reminder to unscreened andnderscreened women identified through a cytology reg-

stry. However, this estimate included the cost of physi-ian participation and laboratory resources to conduct
nd process the additional Pap tests resulting from the a
uly 2008

ntervention. These costs are not typically assigned asntervention costs, and their contribution to overallosts could not be determined by Community Guide staff.hus, the last estimate likely represents an unspecified

nflation of the true cost per additional cancer screen-ng for this intervention.

olorectal Cancer

ffectiveness. Seven studies24,25,43,48,77–79 of greatestesign suitability were identified that reported usinglient reminders to increase colorectal cancer screen-ng by guaiac-based FOBT; no studies were found ofcreening by flexible sigmoidoscopy, colonoscopy, orarium enema. Two studies24,25 were excluded due to

imited quality of execution and one43 was excludedecause the comparison group also received a re-inder, although different from that of the study

roup. The remaining four qualifying studies,48,77–79 allad fair quality of execution.Each study enrolled eligible men and women due or

verdue for annual FOBT, and confirmed test comple-ion by record reviews. The four qualifying studiesvaluated eight intervention arms. One study77 evalu-ted three interventions, two studies78,79 each evaluatedwo interventions, and one study48 evaluated one inter-ention. Three48,77,79 interventions included printedeminders alone; one77 intervention assessed printedeminders with follow-up telephone reminders; andour77–79 interventions included telephone reminderslone or following distribution of an informationalooklet.78 None of the interventions were tailored toddress individual-specific barriers to screening.

Effects of all eight interventions were in the favorableirection, with a median post-intervention increase of1.5 percentage points (IQI�8.9, 20.3; Figure 4). Theagnitude of this effect and the consistent positive

esults across studies and reminder systems demon-trate that client reminders are effective in increasingolorectal cancer screening by FOBT. There were tooew studies to evaluate effect size by type of reminder,ither across or within studies.

pplicability. The same body of evidence was used tovaluate the applicability of client reminders for in-reasing colorectal cancer screening by FOBT in differ-nt settings and populations. Effective interventiontudies that specified settings were conducted in HMOsn the U.S. and in clinics in Canada and Israel. Thesetudies offered limited or no additional description ofhe socioeconomic, racial, ethnic, or screening back-rounds of study participants, or of the geographicettings in which the studies were conducted. However,iven related bodies of evidence on reminders forreast and cervical cancer, client reminders for colo-ectal cancer screening by FOBT should be applicablever a range of settings and populations, provided they
re adapted to target populations and delivery context.

Tsia

EriFraep$ai

C

AsaPaeptbswh

iFwccc

R

Ccsfcice

B

Eobdt

lbae

C

Aicbq

R

Mme

Fn

S

hese findings would not apply to screening by flexibleigmoidoscopy, colonoscopy, or double contrast bar-um enema, because none of the qualifying studiesddressed these procedures.

conomic efficiency. One study77 met inclusion crite-ia8 for cost-effectiveness analysis of client reminders inncreasing colorectal cancer screening by guaiac-basedOBT. Reminders included a postcard alone, a phoneeminder alone, or a combination of the two. Costs perdditional screening, calculated by Community Guideconomists from reported data, were $6.17 for theostcard, $55.41 for the telephone reminder, and38.30 for the combination. The reminder postcardlone therefore appeared to be most cost-effective forncreasing colorectal cancer screening in this study.

onclusions About Client Reminders

ccording to Community Guide methods,16 there istrong evidence that client reminders increase breastnd cervical cancer screening by mammography andap test, respectively. These findings should applycross a range of settings and populations. Althoughvidence also suggests that enhancement of simplerinted reminders with additional messages or supporto clients results in greater effectiveness, particularly forreast cancer screening, it is not yet known whetheruch enhancement increases effectiveness amongomen who have never been screened or who may beard to reach.There is sufficient evidence that client reminders

ncrease colorectal cancer screening by guaiac-basedOBT. Evidence is insufficient, however, to determinehether client reminders are effective in increasingolorectal cancer screening by flexible sigmoidoscopy,olonoscopy, or double contrast barium enema, be-

igure 4. Percentage point change in completed FOBT screenonsignificant

ause no qualifying studies addressed these procedures. n


esults: Client Incentives

lient incentives are small, noncoercive rewards (e.g.,ash or coupons) to motivate people to seek cancercreening for themselves or to encourage others (e.g.,amily members, close friends) to seek screening. In-entives are distinct from interventions designed tomprove access to services (e.g., transportation, childare, reducing out-of-pocket client costs), reviewedlsewhere in this supplement.13

reast, Cervical, and Colorectal Cancers

ffectiveness. No studies were found that reported usef client incentives alone to increase screening forreast, cervical, or colorectal cancers. Therefore, evi-ence was insufficient to determine the effectiveness ofhis intervention when used alone.

Because intervention effectiveness was not estab-ished, the general applicability of client incentives inreast, cervical, or colorectal cancer screening was notddressed, nor was a search made for evidence ofconomic efficiency.

onclusions About Client Incentives

ccording to Community Guide methods,16 there isnsufficient evidence to determine the effectiveness oflient incentives alone in increasing screening forreast, cervical, or colorectal cancer, because no studiesualified for review.

esults: Mass Media

ass media—including television, radio, newspapers,agazines, and billboards—are used to communicate

ducational and motivational information in commu-

ttributable to client reminders. IQI, interquartile interval; NS,
ing a
ity or larger-scale intervention campaigns. Mass media


icvvamgmcbttpcoimicnmnw

B

Eust

ectwogc

aa2cftu

lawo

C

Aima

ro

R

SlTooisfsrsbmpau

B

Erscqwatoolac

covbppptvtscsts(i

J

nterventions, however, almost always include otheromponents or attempt to capitalize on existing inter-entions and infrastructure. For example, such inter-entions have been shown to be effective in reducinglcohol-related motor vehicle crashes when imple-ented with ongoing law enforcement and other pro-

rams to reduce drinking and driving.80 More com-only, mass media occupy a prominent role in broader

ommunity-wide campaigns in which they are com-ined with one or more other components; such mul-icomponent mass media campaigns have been foundo be effective in promoting child safety seat use81 andhysical activity,82 and preventing or reducing adoles-ent tobacco use.83 Small-media messages and group orne-on-one education (see below) are often included

n these programs, based on the rationale that use ofass media to raise awareness and increase knowledge

s most effective in changing health behaviors whenombined with messages delivered through other chan-els.84 At the same time, the individual contribution ofass media to the effectiveness of these multicompo-ent interventions and the effectiveness of mass media,hen used alone, remain uncertain.

reast, Cervical, and Colorectal Cancers

ffectiveness. No studies were found of mass mediased alone to increase breast or colorectal cancercreening. Therefore, evidence was insufficient to de-ermine effectiveness.

Three studies85–87 were identified that evaluated theffectiveness of mass media alone in increasing cervicalancer screening (one study85 repeated the interven-ion in two separate geographic areas). One study 87

as excluded due to limited quality of execution. Thethers, both with fair quality of execution, were ofreatest85 and least86 suitable study design and in-luded three intervention arms.

Post-intervention changes in Pap test completionscertained by record reviews were reported as 20.4%nd 47.6% (relative) increases in one study85 and1.3% in the other.86 These measures could not beonverted to absolute change. Because there were tooew studies of adequate quality, evidence is insufficiento determine the effectiveness of mass media whensed alone in increasing cervical cancer screening.Because intervention effectiveness was not estab-

ished, the general applicability of mass media uselone in breast, cervical, or colorectal cancer screeningas not addressed, nor was a search made for evidencef economic efficiency.

onclusions About Mass Media

ccording to Community Guide methods,16 there isnsufficient evidence to determine the effectiveness of

ass media alone in increasing screening for breast
nd colorectal cancer because no studies qualified for c
uly 2008

eview, and for cervical cancer because too few studiesf adequate quality qualified for review.

esults: Small Media

mall media include videos or printed materials (e.g.,etters, brochures, pamphlets, flyers, or newsletters).hese can be distributed from healthcare systems orther community settings, and can convey educationalr motivational information to promote cancer screen-

ng in target populations. Messages may describecreening tests and procedures and include indicationsor, benefits of, and ways to overcome barriers tocreening. Messages are often based on behavior theo-ies which posit that change in predisposing factors,uch as attitudes, beliefs, and knowledge, are necessaryut not sufficient to achieve behavior changes.2,84 Theseessages may be untailored to address a general target

opulation or tailored to address unique psychologicalnd behavioral characteristics identified through individ-al assessments.88

reast Cancer

ffectiveness. Twenty studies89–108 were identified thateported using small media to increase breast cancercreening by mammography. One108 study was ex-luded due to limited quality of execution. Of 19ualifying studies, 17 had greatest design suitability, ofhich three90,92,98 had good quality of executionnd 1489,91,93–96,99–101,103–107 had fair quality of execu-ion. Two qualifying studies, one with moderate97 andne with least102 suitable study design, had fair qualityf execution. Five studies90,92,100,105,107 evaluated tai-

ored interventions, twelve89,91,93–98,101,102,104,106 evalu-ted untailored interventions, and two studies99,103 in-luded both a tailored and an untailored intervention.

Qualifying studies measured post-interventionhange in completed mammography basedn self-reports89–92,97,100,103–105,107 or record re-iews.93–96,98,99,101,102,106 Tailored interventions usedooklets,90,100 personalized letters,92,99,107 or otherrinted materials.105 Untailored interventions usedersonal checklists or record-keeping booklets,89,94,95

rinted information distributed at medical facili-ies,93,98,103 informational or motivational posters andideos in patient waiting102 or other104 areas, let-ers,91,97,99,101,106 or a video with brochures.93 Onetudy104 was excluded from analysis because it onlyompared three variations of an informational slidehow, and one96 was excluded because it only reportedhe outcome measure as “not significant.” The 17tudies analyzed included 21 intervention armsfour studies93,99,101,103 each evaluated two separatenterventions).

Overall, the median post-intervention increase in
ompleted mammography was 7.0 percentage points

(aob

((tltwmitac

AemEKAwbetitertwi

Ei

cs

C

EicwOseqsi

cslvpotfvrs

Ppnamtv

FN

S

IQI�0.3, 13.2; Figure 5). The magnitude of this effectnd the consistent positive results across studies dem-nstrate the effectiveness of small media in increasingreast cancer screening by mammography.Median increases for tailored (n�7) and untailored

n�14) small media were 7.0 (IQI��4.5, 11.2) and 4.7IQI�0.5, 13.4) percentage points, respectively. Onlywo studies directly compared a tailored and an untai-ored intervention arm. In one,103 the tailored interven-ion was more effective by 5.1 percentage points,hereas in the other,99 the untailored intervention wasore effective by 17.5 percentage points. Overall, stud-

es in this review did not report sufficient informationo examine how the relative effectiveness of tailorednd untailored interventions may vary by populationharacteristics.

pplicability. The same body of evidence was used tovaluate the applicability of small media in increasingammography in a variety of settings and populations.ffectiveness was demonstrated in Australia, the Unitedingdom, and in diverse U.S. populations—includingfrican Americans, whites, and Hispanics—and amongomen of low SES. Effectiveness was also shown inoth rural and urban settings. Findings should gen-rally apply to both tailored and untailored interven-ions across a range of populations, provided thentervention program is appropriately adapted to thearget population and delivery context. The body ofvidence was not large enough to determine theelative effectiveness and merits of tailored and un-ailored approaches overall or whether individualsith certain characteristics benefit from tailored

nterventions.

conomic efficiency. No studies were found meeting

igure 5. Percentage point change in completed mammogramS, nonsignificant

nclusion criteria for review of the economic effi- t


iency of small media in increasing breast cancercreening.

ervical Cancer

ffectiveness. Fourteen studies94,95,99,102,108–117 weredentified that reported using small media to increaseervical cancer screening by Pap test. Two studies108,111

ere excluded because of limited quality of execution.f the 12 qualifying studies, 11 had greatest design

uitability, of which two113,115 had good quality ofxecution and nine94,95,99,109,110,112,114,116,117 had fairuality of execution. One102 study had a least suitabletudy design and fair quality of execution. Two stud-es99,115 included a tailored intervention.

Qualifying studies measured post-intervention change inompleted Pap tests based on record reviews. Twotudies94,95 used personal checklists or record book-ets to inform and prompt participants; two usedideos in patient waiting areas with102 or without117

osters; seven used mailed leaflets,113 brochures,112

r letters;99,109,110,114,115 and one116 used a combina-ion of mailed information in printed and videoormat. The 12 qualifying studies included 15 inter-ention arms (two studies99,109 evaluated two sepa-ate interventions and one study117 conducted theame intervention in two separate clinics).

Overall, the median post-intervention increase inap test completion for 12 intervention arms was 4.5ercentage points (IQI�0.2, 9.0; Figure 6). The mag-itude of this effect and consistent positive resultscross studies demonstrate the effectiveness of smalledia in increasing cervical cancer screening by Pap

est. Effectiveness is further supported by three inter-ention arms109,110 showing statistically significant rela-

ening attributable to small media. IQI, interquartile interval;
scre
ive increases in screening completion, findings not


ic

sttrmpt

Aeivdticitetcu

Eftc

C

Etcsbe

rhc

crppoonaii2ctcscnc

AevdttwicaFitt

Fn

J

ncluded in the analysis because they could not beonverted to percentage point change.

In the only studies of tailored interventions, one99

uggested the tailored intervention (personalized let-er) was 22.2 and 16.2 percentage points less effectivehan an untailored intervention and no intervention,espectively; the other115 suggested that tailored printaterial was no more effective (i.e., 2 percentage

oints less) than the usual practice offered relative tohe comparison group.

pplicability. The same body of evidence was used tovaluate the applicability of small-media interventionsn increasing cervical cancer screening by Pap test in aariety of settings and populations. Effectiveness wasemonstrated in Australia and in diverse U.S. popula-ions, including African Americans, whites, and Hispan-cs, and women of low SES. Findings should be appli-able across a range of populations, provided thentervention program is appropriately adapted to thearget population and delivery context. There was notnough information to generalize the findings abouthe effectiveness of tailored interventions in increasingervical cancer screening or to ascertain which individ-als benefit from tailored interventions.

conomic efficiency. One study,109 classified as satis-actory, estimated the cost of an untailored informa-ional letter to be $15.82 per additional Pap testompleted.

olorectal Cancer

ffectiveness. Nine studies94,108,118–124 were identifiedhat reported using small media to increase colorectalancer screening by guaiac-based FOBT; no studies ofcreening by flexible sigmoidoscopy, colonoscopy, orarium enema were found. Two studies108,122 were

igure 6. Percentage point change in completed Pap test scronsignificant

xcluded because of limited quality of execution. The t

uly 2008

emaining seven94,118–121,123,124 qualifying studies eachad greatest suitability of design and fair quality of exe-ution. One study120 evaluated a tailored intervention.

Qualifying studies measured post-interventionhange in completed FOBT based on self-reports120 orecord reviews.94,118,119,121,123,124 One study94 used aersonal record booklet; three118,119,124 used leaflets oramphlets; two120,123 mailed out videos, newsletters, orther printed materials; and one121 mailed a sequencef two letters. The seven qualifying studies includedine intervention arms (two studies123,124 each evalu-ted two separate interventions). The median post-ntervention increase in completed FOBT for eightntervention arms was 12.7 percentage points (IQI�0,6.4; Figure 7). The magnitude of this effect and theonsistent positive results across studies demonstratehe effectiveness of small media in increasing colorectalancer screening by FOBT. Effectiveness is furtherupported by an intervention arm118 showing a statisti-ally significant OR in the favorable direction, a findingot included in the analysis because it could not beonverted to percentage point change.

pplicability. The same body of evidence was used tovaluate the applicability of these interventions in aariety of settings and populations. Effectiveness wasemonstrated in studies in the United Kingdom andhe U.S., among African-American and white popula-ions, and among some low-SES populations. Studiesere conducted in urban and rural populations and

ncluded study participants from both clinical andommunity settings. It is likely that the findings arepplicable in increasing colorectal cancer screening byOBT across a range of populations, provided the

ntervention program is adapted to the target popula-ion and delivery context. These findings do not applyo flexible sigmoidoscopy, colonoscopy, or double con-

g attributable to small media. IQI, interquartile interval; NS,
eenin
rast barium enema, because no qualifying studies


aoeuw

EogaaG

C

AsvpTadicbdtqtw

R

Gfsmi

ppiibbga

B

ErcwosfiOftb

cratm2fgtttcg

Fn

S

ddressed these procedures. Because there was onlyne tailored intervention study, it was not possible tovaluate the relative effectiveness of tailored versusntailored efforts to increase FOBT or to ascertainhich individuals benefit from tailored interventions.

conomic efficiency. One study120 estimated the costf producing a series of tailored newsletters and tar-eted videos for delivery to 76 people to be $141.60 perdditional guaiac-based FOBT completed. Costs perdditional screening were calculated by Communityuide staff from reported data.

onclusions About Small Media

ccording to Community Guide methods,16 there istrong evidence that small media increase breast, cer-ical, and colorectal cancer screening by mammogra-hy, Pap test, and guaiac-based FOBT, respectively.hese findings should apply across a range of settingsnd populations. Evidence, however, is insufficient toetermine whether small media is effective in increas-

ng colorectal cancer screening by flexible sigmoidos-opy, colonoscopy, or double contrast barium enema,ecause no qualifying studies addressed these proce-ures. For breast cancer screening, the findings applyo both tailored and untailored interventions, althoughuestions remain about the relative effectiveness ofailored versus untailored small-media messages andhich individuals benefit from tailored interventions.

esults: Group Education

roup education conveys information on indicationsor, benefits of, and ways to overcome barriers tocreening with the goal of informing, encouraging, andotivating participants to seek recommended screen-

igure 7. Percentage point change in completed FOBT screonsignificant

ng. Group education is usually conducted by health s


rofessionals or by trained laypeople who use slideresentations or other teaching aids in a lecture or

nteractive format, and often incorporate role model-ng or other methods.84 Because group education cane given to a variety of groups, in different settings, andy different types of educators with different back-rounds and styles, it has been difficult to generalizebout the effectiveness of these interventions.

reast Cancer

ffectiveness. Nine studies125–133 were identified thateported using group education to increase breastancer screening by mammography. Two studies125,126

ere excluded due to limited quality of execution. Sixf the seven qualifying studies were of greatest designuitability, one130 having good quality of execution andve127,129,131–133 having fair quality of execution.ne128 qualifying study had a least suitable design and

air quality of execution. Interventions offered educa-ional information in lecture or interactive format oroth.The seven qualifying studies evaluated post-intervention

hange in mammography completion based on self-eports. These studies evaluated eight interventionrms (one127 study conducted two separate interven-ions). Overall, the median post-intervention change in

ammography was 9.0 percentage points (IQI�4.0,4.0). However, because findings were inconclusiverom four of the eight study arms (the only study withreatest design suitability and good quality of execu-ion130 reported a –1.0 percentage point change andhree129,131,133 measures, with changes ranging from 3.0o 7.0 percentage points, were not statistically signifi-ant), evidence was insufficient to determine whetherroup education is effective in increasing breast cancer

attributable to small media. IQI, interquartile interval; NS,
ening
creening.


lbs

C

EtcetstAa

cO(toic

lcs

C

Eficsedaoadapfidi

lca

C

Aigcc

ito

R

Oubwishahasbaec

B

Ewtrodee

efiiittssvtas

c9fbt

vvib

J

Because intervention effectiveness was not estab-ished, the general applicability of group education inreast cancer screening was not addressed, nor was aearch made for evidence of economic efficiency.

ervical Cancer

ffectiveness. Four studies125,129,133,134 were identifiedhat reported using group education to increase cervi-al cancer screening by Pap test. Two studies125,134 werexcluded from review because of limited quality of execu-ion. Both qualifying studies129,133 were of greatest designuitability and had fair quality of execution. The interven-ions were culturally targeted to Hispanic129 and Filipino-merican133 women and were offered in an interactivend a combined lecture-interactive format, respectively.

Both qualifying studies evaluated post-interventionhange in Pap test completion based on self-reports.ne129 reported an increase of 9 percentage points

p�0.05) and the other133 reported no change followinghe culturally targeted interventions. Because there werenly two studies with inconsistent findings, evidence was

nsufficient to determine the effectiveness of group edu-ation in increasing cervical cancer screening.

Because intervention effectiveness was not estab-ished, the general applicability of group education inervical cancer screening was not addressed, nor was aearch made for evidence of economic efficiency.

olorectal Cancer

ffectiveness. A single multi-arm study135 was identi-ed that reported using group education to increaseolorectal cancer screening by guaiac-based FOBT. Thetudy, of greatest design suitability and good quality ofxecution, included three didactic interventions con-ucted at communal meal sites for older citizens (meange, 72 years). Each was evaluated based on the numberf returned FOBT kits. Two study arms using differentpproaches resulted in 5.0 and �13.0 percentage pointifferences (p�0.05 for both) in kits returned and onerm combining both approaches resulted in a 37ercentage point change (p�0.05). Because intrastudyndings were inconsistent, evidence was insufficient toetermine effectiveness of group education in increas-

ng colorectal cancer screening by FOBT.Because intervention effectiveness was not estab-

ished, the general applicability of group education inolorectal cancer screening was not addressed, nor wassearch made for evidence of economic efficiency.

onclusions About Group Education

ccording to Community Guide methods,16 there isnsufficient evidence to determine the effectiveness ofroup education in increasing screening for breast,ervical, and colorectal cancer. This was due to incon-
lusive findings for breast cancer; too few studies with p
uly 2008

nconsistent findings for cervical cancer; and inconsis-ent findings among multiple intervention arms in thenly study for colorectal cancer.

esults: One-on-One Education

ne-on-one education conveys information to individ-als by telephone or in person on indications for,enefits of, and ways to overcome barriers to screeningith the goal of informing, encouraging, and motivat-

ng them to seek recommended screening. These mes-ages are delivered by healthcare workers or otherealth professionals, lay health advisors, or volunteers,nd are conducted in medical, community, worksite, orousehold settings. Interventions can be untailored toddress a general target population or tailored to reachpecific individuals based on unique psychological andehavioral characteristics as derived from individualssessments.88 As defined for this review, one-on-oneducation may be accompanied by a small-media or alient reminder component.

reast Cancer

ffectiveness. Twenty-six studies41,54,57,91,92,100,105,136–154

ere identified that reported using one-on-one educa-ion to increase breast cancer screening by mammog-aphy. One study136 was excluded due to limited qualityf execution. All 25 qualifying studies were of greatestesign suitability; five41,92,142,147,154 had good quality ofxecution and the remaining 20 had fair quality ofxecution.Qualifying studies evaluated completed mammography,

ither self-reported54,57,91,92,100,105,137–143,145–150,153 or con-rmed by record reviews.41,144,151,152,154 These 25 stud-

es evaluated 35 intervention arms: 14 stud-es54,91,92,100,105,140–143,145,146,148,149,154 evaluated 17ailored intervention arms (two studies91,142 evaluatedwo intervention arms, and one study143 evaluated theame intervention in two distinct populations); eighttudies41,57,137,144,150–153 evaluated ten untailored inter-ention arms (two studies41,57 evaluated two interven-ion arms); two studies138,139 each evaluated two tailoredrms and one untailored intervention arm; and onetudy147 evaluated one tailored and one untailored arm.

Overall, the median post-intervention increase inompleted mammography for 31 intervention arms was.3 percentage points (IQI�4.9, 15.0; Figure 8). ORsrom four additional intervention arms139,146 could note converted to percentage point change but were inhe favorable direction.

Effectiveness did not vary by intervention site (medicalersus nonmedical setting), mode of delivery (telephoneersus in person), or training status of the person deliver-ng the intervention. Overall, there was no differenceetween tailored (n�19, median increase 9.3 percentage
oints) and untailored interventions (n�12, median in-

citwapoet

AesUcaHpsitpdeev

Esob

gscppsitsceecstttaawccc

C

Eii

Ft

S

rease 10.6 percentage points). Of three studies withntrastudy comparisons, however, two138,139 demonstratedhat both tailored interventions in the respective studyere more effective than an untailored intervention (11.0nd 9.0 percentage point increases vs 2.0 percentageoints in one138 and ORs of 2.3 and 2.0 vs 1.5 in thether139), and the third147 also demonstrated greaterffectiveness in the tailored than the untailored interven-ion (9.3 vs �3.0 percentage points).

pplicability. The same body of evidence was used tovaluate the applicability of these findings in a variety ofettings and populations. Studies were conducted in the.S., the United Kingdom, Spain, and Singapore, in

linical and nonclinical (including household) and urbannd rural settings, and included African-American, white,ispanic, and Asian women with and without recent orrevious history of screening mammography. Findingshould generally apply to both tailored and untailorednterventions across a range of populations, providedhe intervention program is adapted to the targetopulation and delivery context. There is some evi-ence from within single studies that tailoring increasesffectiveness. At the same time, however, the body ofvidence is not large enough to ascertain which indi-iduals benefit from tailored interventions.

conomic efficiency. Three studies57,155,156 met inclu-ion criteria8 for analysis of cost effectiveness of one-n-one education in increasing breast cancer screening

igure 8. Percentage point change in completed mammograile interval; NS, nonsignificant

y mammography. Two studies57,155 were classified as s


ood and one156 was classified as very good. Onetudy57 used only direct variable costs and reportedost-effectiveness estimates of $60.47 per additionalerson screened for one-on-one education delivered inerson and $99.37 for delivery by telephone. Twotudies assessed costs from a societal perspective andncluded fixed costs (use of facility and other infrastruc-ure resources) related to the mammography visit it-elf.155,156 One of these studies155 reported a $103.99ost-effectiveness estimate for telephone message deliv-ry without a motivational component and a $156.55stimate for the same intervention with a motivationalomponent. The motivational intervention was bothlightly less effective and substantially more costly thanhe nonmotivational intervention. The other study es-imated cost effectiveness of tailored, barrier-specificelephone counseling at $216, $559, or $1036 perdditional mammogram, depending on whether labornd other resources were volunteered and donated,hether they were offered at minimum wage and actualost, or whether they were based on prevailing wage andommercial cost, respectively.156 This study also estimatedost per life year saved at $14,532 for the intervention.

ervical Cancer

ffectiveness. Eight studies64,113,153,154,157–160 weredentified that reported using one-on-one education toncrease cervical cancer screening by Pap test. Three

eening attributable to one-on-one education. IQI, interquar-
m scr
tudies64,157,158 were excluded due to limited quality of


edt

sreaiTcamruc

iue

Aevfahwptptei

Ei

os

C

Eptqqi

ivtdaaapqiti

ltne

C

Asba

Fi

J

xecution. All five qualifying studies were of greatestesign suitability; two113,154 had good quality of execu-

ion and three153,159,160 had fair quality of execution.Qualifying studies evaluated completed Pap test

creening using either self-reports154,159,160 or recordeviews.113,153 The five qualifying studies evaluatedight intervention arms. Three studies154,159,160 evalu-ted five tailored intervention arms and two stud-es113,153 evaluated three untailored intervention arms.he median post-intervention increase in Pap testompletion over the eight study arms was 8.1 percent-ge points (IQI�5.7, 17.3; Figure 9). Overall, theagnitude of this effect and the consistent positive

esults across studies demonstrate that one-on-one ed-cation interventions are effective in increasing cervicalancer screening by Pap test.

There were too few interstudy measures (and nontrastudy comparisons) to distinguish effectiveness byse of tailoring, or by intervention site, mode of deliv-ry, or training status of the interventionist.

pplicability. The same body of evidence was used tovaluate the applicability of these interventions in aariety of settings and populations. Studies evaluatedace-to-face interventions and were conducted in the U.S.nd the United Kindom; in urban and rural, mostlyousehold, settings; and included African-American,hite, Native American, and Asian, but not Hispanicopulations. Findings should generally apply to bothailored and untailored interventions across a range ofopulations, provided the interventions are adapted tohe target population and delivery context. The body ofvidence was not large enough to ascertain whichndividuals benefit from tailored interventions.

conomic efficiency. No studies were found meeting

igure 9. Percentage point change in completed Pap test scnterval; NS, nonsignificant

nclusion criteria for review of the economic efficiency t

uly 2008

f one-on-one education in increasing cervical cancercreening.

olorectal Cancer

ffectiveness. Two studies77,78 were identified that re-orted using one-on-one education to increase colorec-al cancer screening by guaiac-based FOBT. Each studyualified for review with greatest design suitability and fairuality of execution. One study78 evaluated a tailored

ntervention.Studies evaluated completed FOBT screening follow-

ng either face-to-face education during routine officeisits77 or telephone follow-up by specially trained in-erventionists after FOBT kits were mailed to clientsue for testing.78 One study included two interventionrms evaluating separate untailored interventions (6.6nd 12.9 percentage point changes, both p�0.05),77

nd the other evaluated a tailored intervention (20.7ercentage point change, p�0.05).78 Because overalluality of execution was not sufficient to support find-

ngs from only two studies, there is insufficient evidenceo determine the effectiveness of one-on-one educationn increasing colorectal cancer screening by FOBT.

Because intervention effectiveness was not estab-ished, the general applicability of one-on-one educa-ion in colorectal cancer screening was not addressed,or was a search made for evidence of economicfficiency.

onclusions About One-on-One Education

ccording to Community Guide methods,16 there istrong evidence that one-on-one education increasesreast and cervical cancer screening by mammographynd Pap test, respectively. These findings, for both

ing attributable to one-on-one education. IQI, interquartile
reen
ailored and untailored interventions, should apply


appsoetlsidoslcw

OI

Nefspbia

PI

Lpccemcmitasmirtmmlniait

nsachvmt

ttpdaelibeimdirsac

Rf

Ffitwsparee

I

Admcocmgr

G

●

S

cross a broad range of settings and populations,rovided the interventions are adapted to the targetopulations and delivery contexts. Although there isome evidence that tailoring enhances effectiveness ofne-on-one education in breast cancer screening, thevidence for any difference between tailored and un-ailored approaches in cervical cancer screening isimited. Moreover, for both breast and cervical cancercreening, evidence was too limited to ascertain whichndividuals benefit from tailored interventions. Evi-ence is insufficient to determine the effectiveness ofne-on-one education in increasing colorectal cancercreening, because too few studies, with methodologicimitations (FOBT), or no studies (flexible sigmoidos-opy, colonoscopy, or double contrast barium enema)ere found.

ther Positive or Negative Effects of Interventions toncrease Community Demand

o reports were found of other positive or negativeffects of interventions to increase community demandor breast, cervical, or colorectal cancer screeningervices on use of other healthcare services (e.g., bloodressure monitoring or adult immunization) or healthehaviors (e.g., on smoking or physical activity) or on

nformed decision making (e.g., in reducing patientutonomy by offering incentives).

otential Barriers to Implementing Interventions toncrease Community Demand

imited resources and infrastructure constitute therimary barriers to implementing interventions to in-rease community demand for breast, cervical, andolorectal cancer screening services. Healthcare deliv-ry systems with limited computer or staffing supportay have difficulty tracking, identifying, and notifying

lients eligible for reminders or recall. Cost may be aajor barrier to obtaining adequate exposure, dose,

ntensity, and quality of mass media campaigns. Accesso effective marketing strategies, educational messages,nd instructional materials (particularly for specificubgroups)—key components of mass media, smalledia, group education, and one-on-one education

nterventions—may be limited by cost and special skillsequired to develop and test these messages. Produc-ion and dissemination of tailored messages (small

edia, one-on-one education, and client reminders)ay be more costly and resource intensive than untai-

ored programs because tailoring generally requiresew data collection (although electronic or other med-

cal records could be used for some simple tailoringlgorithms), development of extensive message librar-es with graphics, and computer programming support
o ensure appropriate individualization. Cost effective-

ess can improve, however, through economies ofcale. In addition, materials libraries (such as thosevailable at the Cancer Control PLANET, http://ancercontrolplanet.cancer.gov) are a potential source ofigh quality, topic- and population-specific messages de-eloped as components of evidence-based programs thatany investigators and public health educators are willing

o share.One additional challenge in implementing existing

ailored interventions in community-based settings ishat the diverse computer programs and computerlatforms upon which tailoring algorithms have beeneveloped and implemented may not be easily adapt-ble to a specific community need or healthcare deliv-ry context. However, as more individuals are able toink to and make use of the Internet, web-based tailoredntervention programs may provide a good solution tooth the cost and complexity of developing and deliv-ring tailored interventions to promote cancer screen-ng. Finally, recruitment training, and support of com-

unity health workers and other interventionists toeliver client reminders and educational messages,

ncluding tailored messages, may pose significant bar-iers in smaller community or free-standing clinicalettings. Regional or other aggregations of populationsnd services might be considered as strategies to over-ome this problem.

esearch Issues for Increasing Community Demandor Screening

or the six intervention approaches, the team identi-ed key research issues that had not been answered in

he review. Researchers are encouraged to considerhich of these questions might be answered as part of

tudies already underway, through studies beinglanned, or through new studies. Research questionsre grouped within each of the two effectivenessatings (i.e., effective based on strong or sufficientvidence or undetermined based on insufficientvidence).

nterventions Shown to Be Effective

dditional evidence of effectiveness. These reviewsemonstrated that three interventions to enhance com-unity demand for breast, cervical, and colorectal

ancer screening—client reminders, small media, andne-one-one education—are effective (strong or suffi-ient evidence) in increasing screening rates for one orore of these cancer sites. However, several important

eneral and specific questions about effectivenessemain.

eneral:

How does the effectiveness of interventions to in-
crease community demand for screening vary with

http://cancercontrolplanet.cancer.gov

http://cancercontrolplanet.cancer.gov

●

●

●

●

●

Cc

●

●

●

Sc

●

●

Oa

●

Eebr

fncpset

●

●

●

Eemoiabrpiief

●

●

●

●

J

the health literacy of a target population orsubpopulation?How can newer methods of communication—including automated telephone calls and Internet-delivered applications—be used to improve delivery,acceptance, and effectiveness of these interventions?How effective are these interventions in increasingscreening by colorectal endoscopy or by doublecontrast barium enema (for which no qualifyingstudies were identified)?What is required to disseminate and implementeffective interventions in community settings acrossthe U.S.?How can or should these approaches be applied toassure that screening, once initiated, is maintainedat recommended intervals?With respect to interventions that may be tailoredto individuals, how are effective tailoring pro-grams adapted, disseminated, and implemented incommunity-based settings across the U.S.?

lient reminders (effective in increasing breast, cervi-al, and colorectal [FOBT only] cancer screening):

Does effectiveness of client reminders for cervicaland colorectal cancer screening vary with use ofsupplemental components, such as follow-upprinted materials, telephone calls, or schedulingassistance intended to overcome barriers toscreening?Can client reminders be adapted or used in con-junction with techniques to reach people whohave never been screened for breast, cervical, orcolorectal cancer or who may be hard to reach forscreening?What is the comparative cost effectiveness of tailoredversus untailored client reminder messages?

mall media (effective in increasing breast, cervical andolorectal [FOBT only] cancer screening):

Does effectiveness of small media differ by choice ofmedium (e.g., letter, video, brochure, or Internet-delivered application), information source (e.g.,personal physician, educator), or intensity or fre-quency of delivery?What is the relative cost effectiveness of tailoredversus untailored messages?

ne-on-one education (effective in increasing breastnd cervical cancer screening only):

What are the minimal and optimal duration, dose,and intensity requirements for one-on-one educa-tional approaches to be effective?

vidence of applicability. Interventions found to beffective by the Task Force were generally examined inroad population segments. However, questions may
emain about their effectiveness in certain settings or
uly 2008

or some populations and population subgroups. It isot practical, nor should it be assumed necessary, toonduct intervention trials in every setting or with everyotential subgroup that might be reached, althoughcience will benefit from the accumulation of suchvaluation data.2 At the same time, several key ques-ions about applicability require attention.

How effective are these interventions in populationsthat already have relatively high screening rates?Are these interventions effective in populations his-torically characterized as hard to reach for screen-ing? If not, can they be adapted to such populations?With respect to tailoring small media, one-on-oneeducation, or client reminders sent to individuals:● Does the increased initial cost of developing

tailored interventions (including the data collec-tion required) versus untailored ones justify tai-loring in promoting cancer screening tests for allscreening procedures and for all populationgroups? Are there economies of scale that mightmake these interventions cost effective when usedin particular settings, such as large HMOs?

● Do style, content, format, and depth of tailoringaffect reception, comprehension, and response tomessages and interventions? Although some stud-ies have been conducted to assess one or more ofthese characteristics, not enough data were avail-able to answer these questions for interventionsreviewed here.

vidence of economic efficiency. Efforts by research-rs to provide complete and detailed economic infor-ation using standard approaches will enhance the

verall value of their contributions and will improventerpretability of cost effectiveness across studies andcross intervention options. Because of the small num-er of studies and their methodologic differences, theeported cost-effectiveness data may not, at this stage,rovide sufficient guidance to decision makers in help-

ng them choose a specific option among a menu ofnterventions. To develop a sound basis for comparativeconomic analyses of cancer screening interventions,uture research should consider:

documenting all cost and effectiveness data ele-ments to enable future sharing and evaluation ofcost-effectiveness for these interventions;clarifying study perspective (i.e., program, client,insurance company, societal) and itemizing costsrelevant to that perspective (i.e., direct, fixed, otherindirect);eliminating from analysis any costs not related to theintervention to enhance screening uptake (e.g.,costs of actual screening tests, diagnosis-related pro-cedures, follow-up treatment); andincluding costs of all intervention components in
multicomponent interventions and, where possible,

potnsotsosh

Etcohac

Erpmnimps

IU

E(fRt

●

●

●

tri

●

●

●

●

ngfitptteWpep

●

●

●

●

●

●

D

TTcrcvpb

S

separating effects of individual components fromoverall intervention effectiveness.

Finally, questions remain about the value to programlanners of cost effectiveness in achieving intermediateutcomes (e.g., additional completed screening) ratherhan the health outcome ultimately desired (e.g., ab-ormal tests detected, life years saved). Further efforts,uch as collection of data to estimate numbers of casesf cancer prevented or detected earlier, resulting fromhe additional completed screening, and measuringubsequent improvement in both quality and quantityf life, are also needed to translate cost effectiveness ofcreening into cost effectiveness of achieving ultimateealth goals.

vidence of other positive or negative effects. Ques-ions concerning whether interventions to increaseommunity demand for cancer screening result inther changes in health behavior or use of otherealthcare services (e.g., blood pressure control ordult immunization) could be included in future can-er screening intervention research.

vidence for barriers to implementation. Additionalesearch is needed to find solutions to barriers thatrevent or limit access to effective interventions, toessages, or to resources for producing and dissemi-ating tailored messages. Research and other inquiry

nto methods for centralizing and cataloging interventionessages, both tailored and untailored, potentially could

rovide avenues of sharing and dialogue and make dis-emination and implementation more feasible.

nterventions for Which Effectiveness Isndetermined

ffectiveness of client incentives (alone), mass mediaalone), and group education has not been establishedor breast, cervical, or colorectal cancer screening.emaining research questions in these areas include

he following:

Are these interventions potentially effective in in-creasing screening of these cancer sites?Are some incentives (e.g., ones of greater cash valueor of greater appeal) more effective than others?Do these interventions result in other positive ornegative changes in healthcare services (e.g., bloodpressure monitoring or adult immunization) orhealth behaviors (e.g., smoking or physical activity)?Could incentives become a barrier to developingroutine recommended screening practices or re-duce patient autonomy in decision making?

Given the inherent expense of mass media interven-ions and costs already expended in efforts to answeremaining questions, it may be prudent to seek answers
n lessons gleaned from studies of other health topics. t

What separate effects, if any, do mass media andother major components contribute to overall effec-tiveness of multicomponent media approaches toincrease screening for breast, cervical, and colorec-tal cancers?What are the minimal and optimal component du-ration, dose, and intensity requirements for theseapproaches to be effective?Does effectiveness differ by mass media channel(e.g., TV, radio, billboard) for a given population orsetting?What combinations of mass media and other inter-ventions are optimal to increase a given cancerscreening behavior or to reach particular targetgroups, such as low-income, ethnic, or minoritypopulations?

It has been difficult to generalize about the effective-ess of group education because of the variety ofroups, settings, educators, and styles. Yet despite insuf-cient evidence of overall effectiveness, group educa-

ion could be effective among selected subsets of theopulation, in certain settings, or under certain condi-ions. Thus, we encourage researchers to address addi-ional basic questions that carefully examine specificlements of group education and target populations.e also encourage voluntary health organizations and

ublic health agencies that remain committed to groupducation to collect additional evaluation data, whereossible, to assess such programs as practiced.

Is group education more effective in some settingsthan in others or when delivered in particular for-mats or by particular kinds of educators?Do some populations benefit more from groupeducation than from other interventions?What are the minimal and optimal number, length,and intensity of group education sessions for interven-tion effectiveness and how does effectiveness vary byscreening site and screening histories of populations?Are there optimal combinations of informationand motivational content within group educationinterventions?Is group education effective when combined withother interventions, such as one-on-one education?What is the cost effectiveness of group education?

iscussion

hese reviews summarize the evidence base that supportsask Force recommendations161 for interventions to in-rease community demand for breast, cervical, and colo-ectal cancer screening services. Interventions to increaseommunity demand are strategically distinct from inter-entions to overcome barriers to access13 or to encourageroviders to deliver these services,14 both of which haveeen reviewed separately. Demand-enhancing interven-
ions concentrate on promoting awareness, knowledge,

arpsphottpbsr

rsaoamnttieitcaprcsfmcssswslMoscacEgobhcni

gi

btatFswesinw

ainmt

Tipdsiur

TPJWrmsS

LO

aD

t

R

J

nd motivation among groups of eligible individuals,anging from those who require only reminding orrompting to those who resist or have not yet consideredcreening. Such interventions may be particularly appro-riate for groups that do not make regular visits toealthcare facilities where screening is recommended orffered, or for groups that lack knowledge or motivationo follow cancer screening recommendations. At the sameime, demand-enhancing interventions cannot achieveotential effectiveness if services are not accessible. Thus,efore specific interventions are chosen, selection oftrategies must be considered in the context of localesources and conditions.

An important limitation of these reviews and theecommendations they support is that they do not offerpecific guidance as to which intervention to select forgiven population or setting, nor do they ensure the

ptimal success of recommended interventions underll circumstances. The choice of one or more recom-ended interventions is likely to be influenced by a

umber of different factors, including overall popula-ion screening rate, location and identity of popula-ions in greatest need, opportunities to deliver specificnterventions, and availability of tracking systems. Forxample, despite general success of client reminders inncreasing rates of cancer screening and other preven-ive services,19–23 they may not be the approach ofhoice in communities where cancer screening ratesre relatively high among those enrolled in healthcarerograms but where identifiable pockets of under-usersequire more targeted or intensive efforts to be edu-ated and motivated. Similarly, when communitycreening rates are relatively high, as may be the caseor cervical cancer,4 interventions that involve costly

ass media or other delivery systems for widespreadoverage may not be optimal for reaching the relativelymall proportion of women who have never beencreened or are not getting screened on recommendedchedules. On the other hand, large scale community-ide interventions may be the appropriate choice when

creening rates are relatively low in the general popu-ation, as in the case of colorectal cancer screening.

aking “the right” selection will rely, to a large degree,n knowledge about local context, culture, needs,creening history, and options for delivery. Any appli-ation of a recommended intervention will need to bedapted to specific target audiences and settings. Can-er Control PLANET (Plan, Link, Act, Network, withvidence-based Tools, http://cancercontrolplanet.cancer.ov/) is an important resource for communities andrganizations seeking to adopt and adapt evidence-ased cancer screening interventions. Its links provideelpful sources of information for determining cancerontrol program priorities, identifying potential part-ers, exploring different intervention approaches, find-

ng and adapting research-tested intervention pro-

uly 2008

rams and products, and planning and evaluating thentervention program.

Two additional limitations of these reviews have eacheen addressed in the Research Issues (above). One ishat they offer little insight into the applicability of thepproaches among specific populations who have inhe past been particularly hard to reach for screening.ew studies specifically addressed effectiveness bycreening history or by social or economic status or,ith the exception of client reminders, noted howffectiveness can be enhanced. There is, however, consen-us that higher intensity delivery, content, or components,ncluding use of more personalized approaches, may beecessary to influence those who are less ready to complyith recommended cancer screening tests.The other limitation is that available studies gener-

lly restricted their focus to post-intervention changesn screening behavior over limited time frames and doot deal with maintenance of screening at recom-ended intervals or ways to optimize effective interven-

ions to sustain screening behaviors, once initiated.5,6

These reviews and the accompanying evidence-basedask Force recommendations161 should be useful in

dentifying and selecting options for cancer screeningromotion interventions when increasing communityemand for these services is indicated. Moreover, re-earch questions provided in this article help to identifymportant gaps in our knowledge base and should besed to guide future research, both in determiningesearch priorities and in allocating research funds.

he authors gratefully acknowledge Barbara Reilley, RN,hD; Robert A. Hiatt, MD, PhD; Prethibha George, MPH; S.ay Smith, MHPA; Mona Saraiya, MD, MPH; and Cornelia

hite, PhD for contributions to one or more interventioneviews described in this paper, including conceptual andethodologic insights offered in preparation for their pre-

entation to the Task Force on Community Preventiveervices.The work of Stephanie Melillo, Nancy Habarta, Kimberly

eeks, and Geetika Kalra was supported with funds from theak Ridge Institute for Scientific Education (ORISE).The findings and conclusions in this paper are those of the

uthors and do not necessarily reflect those of the Centers forisease Control and Prevention.No financial disclosures were reported by the authors of

his paper.

eferences1. Centers for Disease Control and Prevention. United States cancer statis-

tics. www.cdc.gov/cancer/npcr/uscs/index.htm.2. National Cancer Policy Board, Institute of Medicine. Fulfilling the poten-

tial of cancer prevention and early detection. Curry S, Byers T, Hewitt M,eds. Washington, DC: The National Academies Press, 2003.

3. Centers for Disease Control and Prevention, National Center for HealthStatistics. Summary health statistics for U.S. adults: National Health Interview
Survey, 2005. www.cdc.gov/nchs/data/series/sr_10/sr10_232.pdf.

http://cancercontrolplanet.cancer.gov/

http://cancercontrolplanet.cancer.gov/

http://www.cdc.gov/cancer/npcr/uscs/index.htm

http://www.cdc.gov/nchs/data/series/sr_10/sr10_232.pdf

S

4. Swan J, Breen N, Coates RJ, Rimer BK, Lee NC. Progress in cancerscreening practices in the United States: results from the 2000 NationalHealth Interview Survey. Cancer 2003;97:1528–40.

5. Clark MA, Rakowski W, Bonacore LB. Repeat mammography: prevalenceestimates and consideration for assessment. Ann Behav Med 2003;26:201–11.

6. Rakowski W, Breen N, Meissner H, et al. Prevalence and correlates ofrepeat mammography among women aged 55-79 in the Year 2000National Health Interview Survey. Prev Med 2004;39:1–10.

7. Breslow RA, Rimer BK, Baron RC, et al. Introducing the CommunityGuide’s reviews of evidence on interventions to increase screening forbreast, cervical, and colorectal cancers. Am J Prev Med 2008;35(1S):S14–S20.

8. Baron RC, Rimer BK, Coates RJ, et al. Methods for conducting systematicreviews of evidence on effectiveness and economic efficiency of interven-tions to increase screening for breast, cervical, and colorectal cancers.Am J Prev Med 2008;35(1S):S26–S33.

9. U.S. Preventive Services Task Force. Guide to clinical preventive services:cancer. www.ahrq.gov/clinic/cps3dix.htm.

10. U.S. Preventive Services Task Force. Screening for breast cancer.www.ahrq.gov/clinic/uspstf/uspsbrca.htm.

11. U.S. Preventive Services Task Force. Screening for colorectal cancer.www.ahrq.gov/clinic/uspstf/uspscolo.htm.

12. U.S. Preventive Services Task Force. Screening for cervical cancer.www.ahrq.gov/clinic/uspstf/uspscerv.htm.

13. Baron RC, Rimer BK, Coates RJ, et al. Client-directed interventions toincrease community access to breast, cervical, and colorectal cancerscreening: a systematic review. Am J Prev Med 2008;35(1S):S56–S66.

14. Sabatino SA, Habarta N, Baron RC, et al. Interventions to increaserecommendation and delivery of screening for breast, cervical, andcolorectal cancers by healthcare providers: systematic reviews of providerassessment and feedback and provider incentives. Am J Prev Med2008;35(1S):S67–S74.

15. Truman BI, Smith-Akin CK, Hinman AR, et al. Developing the Guide toCommunity Preventive Services—overview and rationale. Am J Prev Med2000;18(1S):18–26.

16. Briss PA, Zaza S, Pappaioanou M, et al. Developing an evidence-basedGuide to Community Preventive Services—methods. Am J Prev Med2000;18(1S):35–43.

17. Zaza S, Wright-De Aguero LK, Briss PA, et al. Data collection instrumentand procedure for systematic reviews in the Guide to Community Preven-tive Services. Am J Prev Med 2000;18(1S):44–74.

18. Jackson N, Waters E, Guidelines for Systematic Reviews in Health Promo-tion and Public Health Taskforce. Criteria for the systematic review ofhealth promotion and public health interventions. Health Promot Int2005;20:367–74.

19. Bonfill X, Marzo M, Pladevall M, Martí J, Emparanza JI. Strategies forincreasing the participation of women in community breast cancerscreening. The Cochrane Library, Issue 3. www.update-software.com/abstracts/ab002943.htm.

20. Briss PA, Rodewald LE, Hinman AR, et al. Reviews of evidence regardinginterventions to improve vaccination coverage in children, adolescents,and adults. Am J Prev Med 2000;18(1S):97–140.

21. Stone EG, Morton SC, Hulscher ME, et al. Interventions that increase useof adult immunization and cancer screening services: a meta-analysis. AnnIntern Med 2002;136:641–51.

22. Tseng DS, Cox E, Plane MB, Hla KM. Efficacy of patient letter reminderson cervical cancer screening: a meta-analysis. J Gen Intern Med2001;16:563–8.

23. Wagner T. The effectiveness of mailed patient reminders on mammogra-phy screenings: a meta-analysis. Am J Prev Med 1998;14(1):64–70.

24. Clementz G, Aldag J, Gladfelter T. A randomized study of cancerscreening in a family practice setting using a recall model. J Fam Pract1990;30:537–41.

25. Ornstein S, Garr D, Jenkins R, Rust P, Arnon A. Computer-generatedphysician and patient reminders: tools to improve population adherenceto selected preventive services. J Fam Pract 1991;32:82–90.

26. Reding D, Lappe K, Krueger M, Kolehouse B, Steneil D, Leer R. Cancerscreening and prevention in rural Wisconsin: the Greater MarshfieldExperience. Wis Med J 1997;96:32–7.

27. Hurley S, Huggins R, Jolley D, Reading D. Recruitment activities andsociodemographic factors that predict attendance at a mammographic
screening program. Am J Public Health 1994;84:1655–8.

28. Richardson A, Williams S, Elwood M, Bahr M. Participation in breastcancer screening: randomized controlled trials of doctors’ letters and oftelephone reminders. Aust J Public Health 1999;18:290–2.

29. Richardson J, Mondrus G, Danley K, Deapen D, Mack T. Impact of amailed intervention on annual mammography and physician breastexaminations among women at high risk of breast cancer. CancerEpidemiol Biomarkers Prev 1996;5:71–6.

30. Schapira D, Kumar N, Clark R, Yag C. Mammography screening creditcard and compliance. Cancer 1992;70:509–12.

31. Turner K, Wilson B, Gilbert F. Improving breast screening uptake:persuading initial non-attenders to attend. J Med Screen 1994;1:199–202.

32. Vietri V, Poskitt S, Slaninka S. Enhancing breast cancer screening in theuniversity setting. Cancer Nurs 1997;20:323–9.

33. Becker D, Gomez E, Kaiser D, Yoshihasi A. Improving preventive care ata medical clinic: how can the patient help? Am J Prev Med 1989;5:353–9.

34. Curry S, Taplin S, Anderman C, Barlow W, McBride C. A randomized trialof the impact of risk assessment and feedback on participation inmammography screening. Prev Med 1993;22:350–60.

35. Davis N, Nash E, Bailey C, Lewis M. Evaluation of three methods forimproving mammography rates in a managed care plan. Am J Prev Med1997;13:298–302.

36. Kendall C, Hailey B. The relative effectiveness of three reminder letters onmaking and keeping mammogram appointments. Behav Med 1993;19:29–34.

37. Meldrum P, Turnbull D, Dobson H, Colquhoun C, Gilmour W, McIlwaineG. Tailored written invitations for second round breast cancer screening:a randomised controlled trial. J Med Screen 1994;1:245–8.

38. Ore L, Hagoel L, Shifroni G, Rennert G. Compliance with mammographyscreening in Israeli women: the impact of a pre-scheduled appointmentand of the letter-style. Isr J Med Sci 1997;33:103–11.

39. Segnan N, Senore C, Giordano L, Ponti A, Ronco G. Promoting partici-pation in a population screening program for breast and cervical cancer:a randomized trial of different strategies. Tumori 1998;84:348–53.

40. Stead M, Wallis M, Wheaton M. Improving uptake in non-attenders ofbreast screening: selective use of second appointment. J Med Screen1998;5:69–72.

41. Taplin SH, Barlow WE, Ludman E, et al. Testing reminder and motiva-tional telephone calls to increase screening mammography: a randomizedstudy. J Natl Cancer Inst 2000;92:233–42.

42. Williams E, Vessey M. Randomised trial of two strategies offering womenmobile screening for breast cancer. Br Med J 1989;299:158–9.

43. Vernon SW, Gilstrap EL, Jackson GL, Hughes JI. An intervention toincrease participation in a work site cancer screening program. HealthValues 1992;16:3–9.

44. Barr JK, Franks AL, Lee NC, Antonucci DM, Rifkind S, Schachter M. Arandomized intervention to improve ongoing participation in mammog-raphy. Am J Manag Care 2001;7:887–94.

45. Bodiya A, Vorias D, Dickson H. Does telephone contact with a physician’soffice staff improve mammogram screening rates? Fam Med1999;31:324–6.

46. Burack R, Gimotty P, George J, Simon M, Dews P, Moncrease A. The effectof patient and physician reminders on use of screening mammography ina health maintenance organization: results of a randomized controlledtrial. Cancer 1996;78:1708–21.

47. Davis N, Lewis M. Evaluation of phone intervention to promote mammog-raphy in a managed care plan. Am J Heath Promot 1997;11:247–9.

48. Hogg W, Bass M, Calonge N, Crouch H. Randomized controlled study ofcustomized preventive medicine reminder letters in a community prac-tice. Can Fam Physician 1998;44:81–8.

49. Landis S, Hulkower S, Pierson S. Enhancing adherence with mammographythrough patient letters and physician prompts. N C Med J 1992;53:575–8.

50. Lantz P, Stencil D, Lippert M, Beversdorf S, Jaros L, Remington P. Breast andcervical cancer screening in a low-income managed care sample: the efficacyof physician letters and phone calls. Am J Public Health 1995;85:834–6.

51. Richards SH, Bankhead C, Peters TJ, et al. Cluster randomised controlledtrial comparing the effectiveness and cost-effectiveness of two primarycare interventions aimed at improving attendance for breast screening.J Med Screen 2001;8:91–8.

52. Somkin C, Hiatt R, Hurley L, Gruskin E, Ackerson L, Larson P. The effectof patient and provider reminders on mammography and Papanicolaousmear screening in a large health maintenance organization. Arch InternMed 1997;157:1658–64.

53. Irwig L, Turnbull D, McMurchie M. A randomized trial of general
practitioner-written invitations to encourage attendance at screeningmammography. Community Health Stud 1990;14:357–64.

http://www.ahrq.gov/clinic/cps3dix.htm

http://www.ahrq.gov/clinic/uspstf/uspsbrca.htm

http://www.ahrq.gov/clinic/uspstf/uspscolo.htm

http://www.ahrq.gov/clinic/uspstf/uspscerv.htm

http://www.update-software.com/abstracts/ab002943.htm

http://www.update-software.com/abstracts/ab002943.htm

1

1

1

1

1

1

1

J

54. King E, Rimer B, Seay J, Balshem A. Promoting mammography usethrough progressive interventions: is it effective? Am J Public Health1994;84:104–6.

55. Mayer J, Bartholomew S, Clapp E, Elder J. Facility-based inreach strategiesto promote annual mammograms. Am J Prev Med 1994;10:353–6.

56. Mohler P. Enhancing compliance with screening mammography recom-mendations: a clinical trial in a primary care office. Fam Med1995;27:117–21.

57. Saywell RM Jr., Champion VL, Zollinger TW, et al. The cost effectivenessof 5 interventions to increase mammography adherence in a managedcare population. Am J Manag Care 2003;9:33–44.

58. Simon MS, Gimotty PA, Moncrease A, Dews P, Burack RC. The effect ofpatient reminders on the use of screening mammography in an urbanhealth department primary care setting. Breast Cancer Res Treat2001;65:63–70.

59. Turnbull D, Irwig L, Adelson P. A randomised trial of invitations to attendfor screening mammography. Aus J Public Health 1991;15:33–6.

60. West DS, Greene P, Pulley L, et al. Stepped-care, community clinicinterventions to promote mammography use among low-income ruralAfrican American women. Health Educ Behav 2004;31(Suppl 4):29S–44S.

61. Vogt TM, Glass A, Glasgow RE, La Chance PA, Lichtenstein E. The safetynet: a cost-effective approach to improving breast and cervical cancerscreening. J Womens Health 2003;12:789–98.

62. Bankhead C, Richards SH, Peters TJ, et al. Improving attendance for breastscreening among recent non-attenders: a randomised controlled trial of twointerventions in primary care. J Med Screen 2001;8:99–105.

63. Bowman J, Sanson-Fisher R, Boyle C, Pope S, Redman S. A randomisedcontrolled trial of strategies to prompt attendance for a Pap smear. J MedScreen 1995;2:211–8.

64. Hunt J, Gless G, Straton J. Pap smear screening at an urban aboriginalhealth service: report of a practice audit and an evaluation of recruitmentstrategies. Aust N Z J Public Health 1998;22:720–5.

65. Palm B, Kant A, Van den Bosch W, Vooijs G, van Weel C. Preliminaryresults of a general practice based call system for cervical cancer screeningin the Netherlands. Br J Gen Pract 1993;43:503–6.

66. Rosser W, McDowell I, Newell C. Use of reminders for preventiveprocedures in family medicine. Can Med Assoc J 1991;145:807–13.

67. Wilson A, Leeming A. Cervical cytology screening: a comparison of twocall systems. Br Med J 1987;295:181–2.

68. Kant A, Palm B, Wentink E, van Weel C. General practitioner basedscreening for cervical cancer: higher participation of women with a higherrisk? J Med Screen 1997;4:35–9.

69. Binstock M, Geiger A, Hackett J, Yao J. Pap smear outreach: a randomizedcontrolled trial in an HMO. Am J Prev Med 1997;13:425–6.

70. Buehler S, Parsons W. Effectiveness of a call/recall system in improvingcompliance with cervical cancer screening: a randomized controlled trial.CMAJ 1997;157:521–6.

71. McDowell I, Newell C, Rosser W. Computerized reminders to encouragecervical screening in family practice. J Fam Pract 1989;28:420–4.

72. Pierce M, Lundy S, Palanisamy A, Winning S, King J. Prospective random-ised controlled trial of methods of call and recall for cervical cytologyscreening. Br Med J 1989;299:160–2.

73. Pritchard D, Straton J, Hyndman J. Cervical screening in general practice.Aust J Public Health 1995;19:167–72.

74. Johnston GM, Boyd CJ, MacIsaac MA, Rhodes JW, Grimshaw RN. Effec-tiveness of letters to Cape Breton women who have not had a recent Papsmear. Chron Dis Can 2003;24:49–56.

75. Burack R, Gimotty P, George J, et al. How reminders given to patients andphysicians affected pap smear use in a health maintenance organization:results of a randomized controlled trial. Cancer 1998;82:2391–400.

76. Hyndman J, Straton J, Pritchard D, Le Sueur H. Cost-effectiveness ofinterventions to promote cervical screening in general practice. Aust N ZJ Public Health 1996;20:272–7.

77. Thompson R, Michnich M, Gray J, Friedlander L, Gilson B. Maximizingcompliance with hemoccult screening for colon cancer in clinical prac-tice. Med Care 1986;24:904–14.

78. Myers R, Ross E, Wolf T, Balshem A, Jepson C, Millner L. Behavioralinterventions to increase adherence in colorectal cancer screening. MedCare 1991;29:1039–50.

79. Vinker S, Nakar S, Rosenberg E, Kitai E. The role of family physicians inincreasing annual fecal occult blood test screening coverage: a prospectiveintervention study. Isr Med Assoc J 2002;4:424–5.

80. Elder RW, Shults RA, Sleet DA, et al. Effectiveness of mass media campaigns
for reducing drinking and driving and alcohol-involved crashes: a systematicreview. Am J Prev Med 2004;27:57–65.
uly 2008

81. Zaza S, Sleet DA, Thompson RS, Sosin DM, Bolen JC, Task Force onCommunity Preventive Services. Reviews of evidence regarding interven-tions to increase use of child safety seats. Am J Prev Med 2001;21(4S):31–47.

82. Kahn EB, Ramsey LT, Brownson R, et al. The effectiveness of interven-tions to increase physical activity: a systematic review. Am J Prev Med2002;22(4S):73–107.

83. Hopkins DP, Briss PA, Ricard CJ, et al. Reviews of evidence regardinginterventions to reduce tobacco use and exposure to environmentaltobacco smoke. Am J Prev Med 2001;20(2S):16–66.

84. Glanz K, Rimer BK, Lewis FM. Health behavior and health education:Theory, research, and practice. San Francisco: Jossey-Bass, 2003.

85. Byles J, Sanson-Fisher R, Redmon S. Effectiveness of three community-based strategies to promote screening for cervical cancer. J Med Screen1994;1:150–8.

86. Howe A, Owen-Smith V, Richardson J. The impact of a television soapopera on the NHS Cervical Screening Programme in the North West ofEngland. J Public Health Med 2002;24:299–304.

87. McCaul KD, Jacobson K, Martinson B. The effects of a state-wide mediacampaign on mammography screening. J Appl Soc Psychol 1998;28:504–15.

88. Kreuter MW, Skinner CS. Tailoring: what’s in a name? Health Educ Res2000;15:1–4.

89. Bastani R, Marcus A, Maxwell A, Das I, Yan K. Evaluation of an interven-tion to increase mammography screening in Los Angeles. Prev Med1994;23:83–90.

90. Bastani R, Maxwell AE, Bradford C, Das IP, Yan KX. Tailored risknotification for women with a family history of breast cancer. Prev Med1999;29:355–64.

91. Champion V, Maraj M, Hui S, et al. Comparison of tailored interventionsto increase mammography screening in nonadherent older women. PrevMed 2003;36:150–8.

92. Champion VL, Skinner CS, Menon U, Seshadri R, Anzalone DC, Rawl SM.Comparisons of tailored mammography interventions at two monthspostintervention. Ann Behav Med 2002;24:211–8.

93. Davis T, Berkel H, Arnold C, Nandy I, Jackson R, Murphy P. Interventionto increase mammography utilization in a public hospital. J Gen InternMed 1998;13:230–3.

94. Dickey L, Petitti D. A patient-held minirecord to promote adult preventivecare. J Fam Pract 1992;34:457–63.

95. Dietrich AJ, Duhamel M. Improving geriatric preventive care through apatient-held checklist. Fam Med 1989;21:195–8.

96. Falvo DR, Tippy PK. Comparison of interventions to increase asymptom-atic women’s use of mammography screening. Health Values1993;17:12–7.

97. Fox SA, Stein JA, Sockloskie RJ, Ory MG. Targeted mailed materials andthe Medicare beneficiary: increasing mammogram screening among theelderly. Am J Public Health 2001;91:55–61.

98. Herman C, Speroff T, Cebul R. Improving compliance with breast cancerscreening in older women: results of a randomized controlled trial. ArchIntern Med 1995;155:717–22.

99. Jibaja-Weiss ML, Volk RJ, Kingery P, Smith QW, Holcomb JD. Tailoredmessages for breast and cervical cancer screening of low-income andminority women using medical records data. Patient Educ Couns2003;50:123–32.

00. Lipkus I, Rimer B, Halabi S, Strigo T. Can tailored interventionsincrease mammography use among HMO women? Am J Gastroenterol2000;18:1–10.

01. McCaul KD, Wold KS. The effects of mailed reminders and tailored messageson mammography screening. J Community Health 2002;27:181–90.

02. Mead V, Rhyne R, Wiese W, Lambert L, Skipper B. Impact of environ-mental patient education on preventive medicine practices. J Fam Pract1995;40:363–9.

03. Rakowski W, Ehrich B, Goldstein M, Rimer B, Pearlman D. Increasingmammography among women aged 40-74 by use of a stage-matched,tailored intervention. Prev Med 1998;27:748–56.

04. Rothman A. Attributions of responsibility and persuasion: increasingmammography utilization among women over 40 with an internallyoriented message. Health Psychol 1993;12:29–47.

05. Rimer BK, Halabi S, Sugg SC, et al. Effects of a mammographydecision-making intervention at 12 and 24 months. Am J Prev Med2002;22:247–57.

06. Seow A, Straughan P, Ng E, Emmanuel S, Tan C, Lee H. Factorsdetermining acceptability of mammography in an Asian population: a
study among women in Singapore. Cancer Causes Control 1997;8:771–9.

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

S

07. Skinner C, Strecher V, Hospers H. Physicians’ recommendations formammography: do tailored messages make a difference? Am J PublicHealth 1994;84:43–9.

08. Turner B, Day S, Borenstein B. A controlled trial to improve delivery ofpreventive care: physician or patient reminders? J Gen Intern Med1989;4:403–9.

09. Byles J, Redman S, Sanson-Fisher R, Boyle C. Effectiveness of twodirect-mail strategies to encourage women to have cervical (Pap) smears.Health Promot Int 1995;10:5–16.

10. Byles J, Sanson-Fisher R. Mass mailing campaigns to promote screeningfor cervical cancer: do they work, and do they continue to work? Aust NZ J Public Health 1996;20:254–60.

11. Del Mar C, Glasziou P, Adkins P, Hua T, Brown M. Do personalised lettersin Vietnamese increase cervical cancer screening among Vietnamesewomen? Aust N Z J Public Health 1998;22:824–5.

12. Eaker S, Adami HO, Granath F, Wilander E, Sparen P. A largepopulation-based randomized controlled trial to increase attendanceat screening for cervical cancer. Cancer Epidemiol Biomarkers Prev2004;13:346 –54.

13. McAvoy B, Raza R. Can health education increase uptake of cervical smeartesting among Asian women? Br Med J 1991;302:833–6.

14. Mitchell H, Hirst S, Cockburn J, Reading D, Staples M, Medley G. Cervicalcancer screening: a comparison of recruitment strategies among olderwomen. Med J Aust 1991;155:79–82.

15. Rimer B, Conaway M, Lyna P, et al. The impact of tailored interven-tions on a community health center population. Patient Educ Couns1999;37:125– 40.

16. Taylor VM, Hislop TG, Jackson JC, et al. A randomized controlled trial ofinterventions to promote cervical cancer screening among Chinesewomen in North America. J Natl Cancer Inst 2002;94:670–7.

17. Yancey A, Tanjasiri S, Klein M, Tunder J. Increased cancer screeningbehavior in women of color by culturally sensitive video exposure. PrevMed 1995;24:142–8.

18. Harris MA, Byles JE, Cockburn J, D’Este C. A general practice-basedrecruitment strategy for colorectal cancer screening. Aust N Z J PublicHealth 2000;24:441–3.

19. Hart A, Barone T, Gay S, Inglis A, Griffin L, Tallon C. The effect oncompliance of a health education leaflet in colorectal cancer screening ingeneral practice in central England. J Epidemiol Community Health1997;51:187–91.

20. Kramish Campbell M, James A, Hudson MA, et al. Improving multiplebehaviors for colorectal cancer prevention among African Americanchurch members. Health Psychol 2004;23:492–502.

21. Lee C. A randomised controlled trial to motivate worksite fecal occultblood testing. Yonsei Med J 1991;32:131–8.

22. Powe B, Weinrich S. An intervention to decrease cancer fatalism amongrural elders. Oncol Nurs Forum 1999;26:583–8.

23. Powe BD. Promoting fecal occult blood testing in rural African Americanwomen. Cancer Pract 2002;10:139–46.

24. Pye G, Christie M, Chamberlain J, Moss S, Hardcastle J. A comparison ofmethods for increasing compliance within a general practitioner basedscreening project for colorectal cancer and the effect on practitionerworkload. J Epidemiol Community Health 1988;42:66–71.

25. Kronenfeld J, Windsor R, Kilgo J, Wichers D. A community healtheducation program on breast and uterine cancer in Alabama. In: HobbsP, ed. Public education about cancer. Geneva: International UnionAgainst Cancer, 1980:8–19.

26. Forsyth M, Fulton D, Lane D, Burg M, Krishna M. Changes in knowledge,attitudes, and behavior of women participating in a community outreacheducation program on breast cancer screening. Patient Educ Couns1992;19:241–50.

27. Aiken L, West S, Woodward C. Increasing screening mammography inasymptomatic women: evaluation of a second-generation, theory-basedprogram. Health Psychol 1994;13:526–38.

28. Erwin D, Spatz T, Ches R, Stotts C, Hollenberg J, Deloney L. Increasingmammography and breast self-examination in African American womenusing the Witness Project Model. J Cancer Educ 1996;11:210–5.

29. Navarro A, Senn K, McNicholas L, Kaplan R, Roppe B, Campo M. Por LaVida model intervention enhances use of cancer screening tests amongLatinas. Am J Prev Med 1998;15:32–41.

30. Mishra S, Chavez L, Magana J, Nava P, Valdez R, Hubbell F. Improvingbreast cancer control among Latinas: evaluation of a theory-based educa-
tional program. Health Educ Behav 1998;25:653–70.

31. King E, Rimer B, Benincasa T, Harrop C, Amfoh K, Bonney G. Strategiesto encourage mammography use among women in senior citizens’housing facilities. J Cancer Educ 1998;13:108–15.

32. Skinner C, Arfken C, Waterman B. Outcomes of the Learn, Share & LiveBreast Cancer Education Program for Older Urban Women. Am J PublicHealth 2000;90:1229–34.

33. Maxwell AE, Bastani R, Vida P, Warda US. Results of a randomized trial toincrease breast and cervical cancer screening among Filipino Americanwomen. Prev Med 2003;37:102–9.

34. Carney P, Dietrich A, Freeman D. Improving future preventive carethrough educational efforts at a women’s community screening program.J Community Health 1992;17:167–74.

35. Weinrich S, Weinrich M, Stromborg M, Boyd M, Weiss H. Using elderlyeducators to increase colorectal cancer screening. Gerontologist1993;33:491–5.

36. Abood DA, Coster DC, Mullis AK, Black DR. Evaluation of a “loss-framed”minimal intervention to increase mammography utilization among medicallyun- and under-insured women. Cancer Detect Prev 2002;26:394–400.

37. Calle E, Moss R, Miracle-McMahill H, Health C. Personal contact fromfriends to increase mammography usage. Am J Prev Med 1994;10:361–6.

38. Champion V. Strategies to increase mammography utilization. Med Care1994;32:118–29.

39. Champion V, Huster G. Effect of interventions on stage of mammographyadoption. J Behav Med 1995;18:169–87.

40. Champion VL, Ray DW, Heilman DK, Springston JK. A tailored interven-tion for mammography among low-income African-American women.J Psychosoc Oncol 2000;18:1–13.

41. Costanza M, Stoddard A, Luckmann R, White M, Avrunin J, Clemow L.Promoting mammography: results of a randomized trial of telephonecounseling and a medical practice intervention. Am J Prev Med2000;19:39–46.

42. Crane L, Leakey T, Rimer B. Effectiveness of a telephone outcallintervention to promote screening mammography. Prev Med 1998;27(5Pt 2):S29–S38.

43. Duan N, Fox S, Derose K, Carson S. Maintaining mammography adher-ence through telephone counseling in a church-based trial. Am J PublicHealth 2000;90:1468–71.

44. Hoare T, Thomas C, Biggs A, Booth M. Can the uptake of breast screeningby Asian women be increased? A randomized controlled trial of link-worker intervention. J Public Health Med 1994;16:179–85.

45. Howze EH, Broyden RR, Impara JC. Using informal caregivers tocommunicate with women about mammography. Health Commun1992;4:227– 44.

46. Lauver D, Kane J. A motivational message, external barriers, and mam-mography utilization. Cancer Detect Prev 1999;23:254–64.

47. Lauver DR, Settersten L, Kane JH, Henriques JB. Tailored messages,external barriers, and women’s utilization of professional breast cancerscreening over time. Cancer 2003;97:2724–35.

48. Marcus A, Bastani R, Reardon K. Proactive screening mammographycounseling within the cancer information service: results from a random-ized trial. J Natl Cancer Inst 1993;14:119–29.

49. Messina CR, Lane DS, Grimson R. Effectiveness of women’s telephonecounseling and physician education to improve mammography screeningamong women who underuse mammography. Ann Behav Med2002;24:279–89.

50. Schwartz MD, Rimer BK, Daly M, Sands C, Lerman C. A randomized trialof breast cancer risk counseling: the impact on self-reported mammogra-phy use. Am J Public Health 1999;89:924–6.

51. Segura JM, Castells X, Casamitjana M, Macia F, Porta M, Katz SJ. Arandomized controlled trial comparing three invitation strategies in abreast cancer screening program. Prev Med 2001;33:325–32.

52. Seow A, Straughan P, Lee H. A randomised trial of the use of printmaterial and personal contact to improve mammography uptake amongscreening non-attenders in Singapore. Ann Acad Med Singapore1998;27:838–42.

53. Sung J, Williams J, Blumenthal D, Alema-Mensah E, Coates R, Liff J. Effectof cancer screening intervention conducted by lay health workers amonginner-city women. Am J Prev Med 1997;13:51–7.

54. Valanis BG, Glasgow RE, Mullooly J, et al. Screening HMO women overduefor both mammograms and pap tests. Prev Med 2002;34:40–50.

55. Fishman P, Taplin S, Meyer D, Barlow W. Cost effectiveness of strategiesto enhance mammography use. Eff Clin Pract 2000;4:213–20.

56. Stockdale S, Keeler E, Duan N, Derose K, Fox S. Costs and cost-
effectiveness of a church-based intervention to promote mammographyscreening. Health Serv Res 2000;35:1037–57.

1

1

1

1

1

J

57. Ward J, Redman S, Boyle J, Sanson-Fisher R. Increasing women’s compli-ance with opportunistic cervical cancer screening: a randomized trial.Am J Prev Med 1991;7:285–91.

58. Ward J, Proude E. Evaluation of doctors’ reminders in emergencydepartments to encourage cervical screening. Aust N Z J Public Health1999;23:95–8.

59. Dignan M, Michielutte R,education to increase scree

uly 2008

Cherokee Indian women in North Carolina. J Natl Cancer Inst1996;88:1670–6.

60. Dignan MB, Michielutte R, Wells HB, et al. Health education to increasescreening for cervical cancer among Lumbee Indian women in NorthCarolina. Health Educ Res 1998;13:545–56.

61. Task Force on Community Preventive Services. Recommendations for
s to increase breast, cervical, anded 2008;35(1S):S21–S25.
Blinson K, et al. Effectiveness of healthning for cervical cancer among eastern-band

client- and provider-directed interventioncolorectal cancer screening. Am J Prev M

Did you know?You can save your online searches and get the results by e-mail.Visit www.ajpm-online.net today to see what else is new online!


Documents

Client-Directed Interventions to Increase Community Demand for