What Is the Value of the Routine Use of Patient-Reported ...

What Is the Value of the Routine Use of Patient-ReportedOutcome Measures Toward Improvement of PatientOutcomes, Processes of Care, and Health Service Outcomesin Cancer Care? A Systematic Review of Controlled TrialsGrigorios Kotronoulas, Nora Kearney, Roma Maguire, Alison Harrow, David Di Domenico, Suzanne Croy,and Stephen MacGillivray

Grigorios Kotronoulas, Nora Kearney,Roma Maguire, University of Surrey,Guildford; Alison Harrow, DundeeCancer Centre; David Di Domenico,Stephen MacGillivray, University ofDundee, Dundee; Suzanne Croy,Dementia Services DevelopmentCentre, University of Stirling, Stirling,United Kingdom.

Published online ahead of print atwww.jco.org on April 7, 2014.

Authors’ disclosures of potential con-flicts of interest and author contribu-tions are found at the end of thisarticle.

Corresponding author: Grigorios Kotro-noulas, PhD, MSc, BSN, School ofHealth and Social Care, University ofSurrey, Duke of Kent Building, Guild-ford, Surrey,GU2 7TE, United Kingdom; e-mail:[email protected].

© 2014 by American Society of ClinicalOncology

0732-183X/14/3214w-1480w/$20.00

DOI: 10.1200/JCO.2013.53.5948

A B S T R A C T

PurposeThe systematic use of patient-reported outcome measures (PROMs) has been advocated as aneffective way to standardize cancer practice. Yet, the question of whether PROMs can lead toactual improvements in the quality of patient care remains under debate. This review examinedwhether inclusion of PROM in routine clinical practice is associated with improvements in patientoutcomes, processes of care, and health service outcomes during active anticancer treatment.

MethodsA systematic review of five electronic databases (Medline, EMBASE, CINAHL [Cumulative Indexto Nursing and Allied Health Literature], PsycINFO, and Psychology and Behavioral SciencesCollection [PBSC]) was conducted from database inception to May 2012 to locate randomized andnonrandomized controlled trials of patients receiving active anticancer treatment or supportivecare irrespective of type of cancer.

ResultsBased on prespecified eligibility criteria, we included 26 articles that reported on 24 uniquecontrolled trials. Wide variability in the design and use of interventions delivered, outcomesevaluated, and cancer- and modality-specific context was apparent. Health service outcomes wereonly scarcely included as end points. Overall, the number of statistically significant findings werelimited and PROMs’ intervention effect sizes were predominantly small-to-moderate.

ConclusionThe routine use of PROMs increases the frequency of discussion of patient outcomes duringconsultations. In some studies, PROMs are associated with improved symptom control, increasedsupportive care measures, and patient satisfaction. Additional effort is required to ensure patientadherence, as well as additional support to clinicians who will respond to patient concerns andissues, with clear system guidelines in place to guide their responses. More research is requiredto support PROM cost-benefit in terms of patient safety, clinician burden, and healthservices usage.

J Clin Oncol 32:1480-1501. © 2014 by American Society of Clinical Oncology

INTRODUCTION

Anticancer treatments have brought about definiteadvances in patient survival rates.1 However, treat-ment is associated with significant toxicity that ispotentially life-threatening,1 and can often result inpoor treatment adherence, impaired quality of life(QoL), and mortality.2,3 Systematic monitoring iscrucial to detect problems, to address needs of pa-tients, and to plan care.4 Using patient-reported out-come measures (PROMs), “measurements of anyaspect of a patient’s health status that come directlyfrom the patient,”5 facilitates a systematic and com-

prehensive approach to patient assessment andidentifies problems that are often overlooked withinroutine practice. Regularly collecting PROM data isan effective way to standardize practice and improvepatient management.4 Nevertheless, the question ofwhether PROMs can improve the quality of patientcare, and whether this relates both to health profes-sional engagement with them and to the systemguidelines in place to guide response, remains underdebate. Given the costs associated with collectingPROMs, evidence of their effect on patient out-comes (POs), processes of care (PoCs), and/orhealth service outcomes (HSOs) is needed.

JOURNAL OF CLINICAL ONCOLOGY R E V I E W A R T I C L E

VOLUME 32 � NUMBER 14 � MAY 10 2014

1480 © 2014 by American Society of Clinical Oncology

Downloaded from ascopubs.org by University of Glasgow Library on February 20, 2018 from 130.209.029.070Copyright © 2018 American Society of Clinical Oncology. All rights reserved.

Previous reviews have concluded some clinically meaningful, butnot always statistically significant, effects on the use of PROMs inclinical practice.5-11 Only two of these reviews9,11 were specific tocancer care and differed in terms of objectives, comprehensiveness,and quality. Taking into consideration the lack of clarity around theuse of PROMs in cancer care, we conducted a comprehensive system-atic review of all available controlled trials (CTs) to examine whetherroutine use of PROMs by health care professionals (HPs) can improvethe quality of care patients receive during active anticancer treatment.The value of PROM use was examined through detection of positiveeffects on POs, PoCs, and HSOs, as suggested by statistical/clini-cal changes.

METHODS

We searched five electronic databases (Medline, EMBASE, CINAHL [Cumu-lative Index to Nursing and Allied Health Literature], PsycINFO, and PBSC)from database inception to May 2012, using a systematic strategy that wasdevised and refined through an iterative process (Appendix Table A1 [online-only]). Additional articles were identified through previous topicalreviews.5-11 We also examined reference lists of the articles retained for anystudies that might have been overlooked. Where necessary, we contacted studyauthors to provide clarification on characteristics of the study samples in-cluded. We followed the Preferred Reporting Items for Systematic Reviews andMeta-Analyses (PRISMA) guidelines where applicable.12

Study Selection Criteria

Trials were deemed eligible if they were primary or secondary reports ofCTs testing PROM interventions in which PROM-generated feedback wasmade available to HPs or patients to improve quality of patient care; involvedadult patients (� 18 years old) with cancer, irrespective of disease stage, whoreceived any type of active anticancer treatment or supportive care, even if onlypart of the sample received active treatment/care but percentages were re-ported; were randomized CTs (RCTs) or non-RCTs; and were published inthe English language with readily available abstracts. Trials were excluded ifthey evaluated PROMs as part of broader psychobehavioral interventions, inwhich PROMs were only used to evaluate intervention effectiveness; investi-gated the effects of a medicinal product; were conducted with survivors ofcancer who were not actively receiving anticancer treatment; tested the psy-chometric properties of PROMs; or involved children with cancer, or survi-vors of childhood cancers.

Study Selection and Data Extraction Procedures

Study selection involved two stages: an initial title and abstract screeningwith eligibility evaluation performed by two screening groups that indepen-dently screened the retrieved records against selection criteria, and retrievingpotentially eligible full-text articles, which were independently evaluated foreligibility by five reviewers. Selection of the final sample of studies was dis-cussed until a consensus was reached. Five reviewers extracted data usingforms that were specifically developed for this review, pilot tested the forms onthree randomly selected studies, and refined the forms accordingly.

Risk of Bias and Methodologic Quality Evaluation

We used the Cochrane Collaboration Risk of Bias Tool13 to evaluate sixdifferent domains of a CT: adequacy of sequence generation, concealment ofallocation, blinding, completeness of follow-up, freedom from reporting bias,and other forms of bias. We evaluated each domain of bias as low risk, highrisk, or unclear. Three reviewers assessed five articles each, and a fourth re-viewer cross-checked the evaluations until a consensus was reached. Reviewerswere not blinded to authors, institutions, or journals of publication.

Outcome Evaluation

Based on previous topical reviews,5-11 three major outcome categorieswere formed: POs (ie, health status/well-being/functioning; symptom burden/distress; health-related QoL; psychological distress), PoCs (ie, patient satisfac-

tion with treatment/care/consultation; patient behaviors/actions/adherence;patient-HP communication; patient-HP concordance in assessments; HP en-gagement in assessment), and HSOs (ie, patient safety; cost-effectiveness;number of contacts with clinicians; patient resources/services use). We antic-ipated that not all CTs would report on every outcome category or on everyoutcome within a specific category.

Synthesis of Results and Determination of Effect Sizes

Individual outcomes were classified according to prespecified majoroutcome categories, and findings were narratively synthesized. Prevalence (%)of studies examining each individual outcome and major categories was ex-amined and plotted. Because of variability in the patient populations, out-comes assessed, outcome PROMs used, and reporting of results, we deemed ameta-analysis was not feasible. However, where enough data were available,effect sizes (ES; Cohen’s d) and 95% CIs were estimated based on meanpostintervention total scores of outcome measures or percentages of patientsreporting specific outcomes based on specific formulas.14,15 By convention, ESwhere d � 0.2 were considered small, d � 0.5 were moderate, and d � 0.8were large.16

RESULTS

Search Results and Study Characteristics

Initial searches retrieved 4,997 references from electronic databases and18 from previous published literature reviews.5-11 Twenty-six articles17-42 re-porting on 24 unique CTs fulfilled eligibility criteria and were included in aqualitative synthesis (Fig 1). All but four trials18,24,34,36 were RCTs, and 16adopted a longitudinal study design (Table 1). Patient study samples variedwidely in size (median, 194 individuals; range, 48 to 1,134 individuals; for atotal of 6,279 individuals). HP samples varied similarly (median, 22 HPs;range, four to 262 HPs; total, n � 713), but they were reported in only 11 trials.Nine CTs tested interventions designed specifically for patients withbreast,20,22,26,27 lung,20,29,30,33,34 or hematologic malignancies.32 SeventeenCTs tested interventions delivered in the outpatient/ambulatory setting. Onlytwo RCTs targeted patients with early-stage cancers.19,22 Thirty-seven percentto 100% of patients were receiving active anticancer treatments duringstudy participation, and these treatments were most frequently chemother-apy or radiotherapy.

In terms of intervention design, patients in the control group eitherreceived usual care only19,21,28,34,36,41,42 or completed PROMs similar to thatof the experimental group, but feedback remained unavailable toHP.17,18,24,26,30-33,37,40 Only one three-arm RCT combined these two alterna-tive conditions in the same design.35,38,39 In the more diverse CTs, PROMswere completed at home by the experimental group but were not administeredto patients in the control group25,29; were completed by all participants, butPROM summaries of the experimental group were only placed in the medicalrecords or sent to HPs20,23; or were completed by patients in the experimentalgroup only to direct further intervention based on distress expressed by asubset of the group.20,22,27 In only five CTs did HPs follow specific guidelinesto guide response to PROM feedback.20,22,23,26,28

Twenty-nine PROMs were administered in the reviewed trials to helpdeliver the interventions (Appendix Table A2). Eleven CTs relied on only oneintervention PROM, seven incorporated two PROMs, and six CTs used threeor more instruments.17,18,23,24,28,42 The most frequently used PROM was theEuropean Organisation for Research and Treatment of Cancer Quality of LifeQuestionnaire C30 (EORTC QLQ-C30; n � 11). Other PROMs focused onsymptom prevalence and severity (n � 11), supportive care needs (n � 8),QoL issues (n � 5), or sources of distress (n � 3). The PROMs were adminis-tered on media including electronic platforms (n � 11), paper-and-penciltools in clinic (n � 12), take-home log books (n � 3), and mailed assessmentsand/or telephone interviews (n � 7; Table 1).

Risk of Bias Within and Across Studies

Two RCTs were rated as low risk in five of the seven bias categories.26,29

Yet, bias in the design and/or reporting was present in all of the included trials(Table 2), regardless of whether patients were randomly assigned to the study

PROMs’ Value in Improving Cancer Care Outcomes

www.jco.org © 2014 by American Society of Clinical Oncology 1481


condition. Only seven RCTs were rated as low risk on both the random-assignment generation process and allocation-concealment bias.17,20,26-29,41

Conversely, all non-RCTs were consistently rated as high risk. With the excep-tion of three RCTs,21,22,40 performance bias was rated as high for all CTs giventhat blinding on the HP level was not feasible. With the exception of sevenCTs,18,25,28-30,40,41 risk of detection bias was also deemed high or unclear. TenCTs were rated as high risk regarding attrition-related bias.18-20,24,27,33,35,38-42

Selective outcome reporting bias was predominantly unclear (n � 18; 75%).Additional sources of bias interfered with 15 CTs. Most frequently, authorswere unclear as to whether HPs who received patient feedback actually used itduring consultations.

Outcomes Evaluation

POs and/or PoCs were reported as primary outcomes in 21 CTs(87.5%) and 19 CTs (79.2%), respectively; however, intervention effects onHSOs were only scarcely investigated (Table 2 and Table 3).20,22,27,30,42

Eighteen CTs evaluated the effects of interventions in the long term (� 8weeks), with follow-up assessments ranging in number from two to four ormore that were conducted for up to 12 months (but mainly � 6 months)after baseline assessment.

Patient Outcomes

Physical symptoms. Overall, positive effects with reduced symptom prev-alence or severity were reported in seven CTs (six RCTs), mainly clinically and lessfrequently statistically significant. ES ranged widely and were mainly small-to-moderate in terms of intervention effects on physical symptom prevalence (d �0.01 to 0.75), physical symptom severity (d�0.0 to 0.44), psychological symptomprevalence (d � 0.07 to 0.15), psychological symptom severity (d � 0.01 to 0.30),or psychological symptom distress (d � 0.09 to 0.42; Appendix Table A3). AcrossCTs, patients in the experimental group reported greater reductions in symptom-threshold events and symptom interference with functioning,40 severity of meno-pausal symptoms and sexual dysfunction,22 frequency of constipation andvomiting,25 incidence of pain37 or fatigue,41 debilitating symptoms,18 and distressassociated with symptoms/problems32,41 compared with those in the controlgroup, irrespective of cancer type or stage.

Quality of life. Survivors of breast cancer,22 patients with nonlocalizedbreast cancer or colorectal cancer,23 and groups of patients with mixed cancer

diagnoses at an advanced stage21,31,42 or at various clinical stages24,28 had nosignificant postintervention effects in nine CTs (Table 2; Appendix Table A3).In terms of overall QoL, ES ranged from 0.04 to 0.59, but were mainly small inmagnitude. Nevertheless, rates of diseased QoL were reduced in women withbreast cancer 6 months after surgery in the experimental group compared withthe control group (d � 0.35).26 Among patients with lung cancer, QoL scoresdeteriorated in the experimental group more than in the standard-care groupover the 16 weeks of observation.29 Velikova et al38 reported improvements inpatient QoL scores at treatment initiation that were influenced by whetherQoL was actually discussed during consultations.38

Psychological symptoms. Results were generally unsupportive of signif-icant postintervention effects on anxiety and/or depression regardless ofwhether direct real-time18,23 or indirect20 patient feedback was made availableto HPs. This was evident despite overall reductions in psychological distressover time.27 Similarly, McLachlan et al28 found no overall intervention effectson depression scores, but the subgroup of patients classified as moderately orseverely depressed benefitted more from the intervention. Where significantimprovements in anxiety or depression were reported,42 these were small-to-moderate in magnitude (d � 0.15 to 0.42) and not universal across all assess-ment PROMs.

Supportive care needs. Five CTs provided generally unclear evidence;despite some small-to-moderate ES (d � 0.16 to 0.58) across domains of need,these were not always in favor of the experimental group (Appendix Table A3).The PROM intervention was no better than usual care in tackling needs ofpatients in two trials.18,23 We found statistically significant between-groupdifferences in 13 of 19 categories of perceived need32 and sexual health con-cerns (d � 0.49)22 in favor of the experimental group among patients withhematologic malignancies32 and breast cancer,22 respectively. In a non-RCT,patients receiving routine psychological screening reported more psychologi-cal, information, and physical/daily living needs, but not sexuality needs, at 6months postbaseline compared with the unscreened cohort.36

Processes of Care

Medical decisions made/advice given/changes in treatment/referrals made.Despite being the outcomes most frequently investigated (Table 3), evidence

Records identified through database searching Medline/EMBASE CINAHL, PsycINFO, PBSC

(n = 1,611)(n = 3,386)

Additional records identified through other sources Previous literature reviews

Records screened(n = 5,015)

Records excluded on basis of title/abstract

(n = 4,967)

Full-text articles assessed for eligibility

(n = 49)

Articles included in narrative synthesis

of findings(n = 26)

Articles identified through reference lists

(n = 1)

(n = 18)

Full-text articles excluded PROMs not the intervention No active treatment Children with cancer No controlled trial No results reported Compared intervention modes

(n = 23)(n = 13)(n = 1)(n = 1)(n = 5)(n = 1)(n = 2)

Fig 1. Diagram of the study selectionprocess according to Preferred Report-ing Items for Systematic Reviews andMeta-Analyses guidelines.12,43 CINAHL,Cumulative Index to Nursing and AlliedHealth Literature; PBSC, Psychology andBehavioral Sciences Collection; PROMs,patient-reported outcome measures.

Kotronoulas et al

1482 © 2014 by American Society of Clinical Oncology JOURNAL OF CLINICAL ONCOLOGY


Tabl

e1.

Sum

mar

ies

ofth

eM

etho

dolo

gic

Cha

ract

eris

tics

ofth

e24

Stu

dies

(26

artic

les)

Rep

ortin

gon

the

Use

ofP

RO

Ms

As

Inte

rven

tions

inP

atie

nts

With

Can

cer

Rec

eivi

ngA

ctiv

eA

ntic

ance

rTr

eatm

ent

Aut

hor

and

Yea

rof

Stu

dyS

ettin

g/Lo

catio

nP

atie

ntP

opul

atio

nTy

peof

Trea

tmen

t�N

o.of

Pat

ient

s†N

o.of

HP

sS

tudy

Des

ign

Inte

rven

tion/

Con

trol

Out

com

esA

sses

sed

Met

hod

ofA

dmin

istr

atio

nof

PR

OM

Eva

luat

ion

ofE

ffec

ts

Pat

ient

Rec

eive

dP

RO

MFe

edba

ck

Ber

ryet

al,17

2011

Out

patie

ntcl

inic

,U

SM

ixed

canc

erdi

agno

ses

(typ

e,st

age,

and

time

sinc

edi

agno

sis)

;st

artin

ga

new

med

ical

orra

diat

ion

trea

tmen

tre

gim

en;

RR

,62

%;

AR

,20

%

47%

MD

;23

%R

T;30

%S

CT

327

(I);

333

(C)

262

Two-

arm

RC

TIn

terv

entio

n:C

ompl

etio

nof

inte

rven

tion

PR

OM

sth

roug

hE

SR

A-C

and

sum

mar

ies

avai

labl

eto

HP

sbe

fore

cons

ulta

tion.

Con

trol

:C

ompl

etio

nof

inte

rven

tion

PR

OM

sth

roug

hE

SR

A-C

but

sum

mar

ies

unav

aila

ble

toH

Ps.

Pro

cess

esof

care

;he

alth

serv

ice

outc

omes

Ele

ctro

nic

inte

ract

ive

tool

Sho

rtte

rm(s

ame

day

asco

nsul

tatio

nvi

sit)

No

Boy

eset

al,18

2006

Out

patie

ntcl

inic

,A

ustr

alia

Mix

edca

ncer

diag

nose

s(t

ype,

stag

e,an

dtim

esi

nce

diag

nosi

s);

atte

ndin

gcl

inic

for

the

first

time;

RR

,65

%;

AR

,40

%

65%

SR

G;

11%

RT;

6%C

T;3%

HT;

4%ot

her

ATR

42(I)

;38

(C)

4P

ilot

long

itudi

nal

two-

arm

non-

RC

T

Inte

rven

tion:

Com

plet

ion

ofto

uch-

scre

enco

mpu

ter

surv

eybe

fore

cons

ulta

tion,

and

sum

mar

ies

avai

labl

eto

cons

ulta

nts.

Con

trol

:C

ompl

etio

nof

touc

h-sc

reen

com

pute

rsu

rvey

befo

reco

nsul

tatio

nbu

tsu

mm

arie

sun

avai

labl

eto

cons

ulta

nts.

Pat

ient

outc

omes

;pr

oces

ses

ofca

re

Ele

ctro

nic

inte

ract

ive

tool

;pa

per

tool

incl

inic

Long

term

(thr

eef/

uvi

sits

)N

o

Bra

eken

etal

,19

2011

Out

patie

ntR

Tcl

inic

,th

eN

ethe

rland

s

Mix

edea

rly-s

tage

canc

erdi

agno

ses;

befo

refir

stco

nsul

tatio

n;sc

hedu

led

tore

ceiv

e�

10fr

actio

nsof

RT;

RR

,51

%;

AR

,N

R

100%

RT

268

(I);

300

(C)

14Tw

o-ar

mcl

uste

red

RC

TIn

terv

entio

n:C

ompl

etio

nof

inte

rven

tion

PR

OM

befo

refir

stan

dla

stco

nsul

tatio

n,an

dre

port

sav

aila

ble

tora

diot

hera

pist

sin

volv

edin

care

;th

era

diot

hera

pist

sdi

scus

sed

patie

ntne

eds

and

refe

rred

patie

nts

tops

ycho

soci

alca

repr

ovid

ers.

Con

trol

:C

are

asus

ual.

Pro

cess

esof

care

Pap

erto

olin

clin

icLo

ngte

rm(t

hree

asse

ssm

ents

afte

rba

selin

ew

ithin

12m

)

No

Car

lson

etal

,20

2010

Out

patie

ntcl

inic

,C

anad

aN

ewdi

agno

sis

ofbr

east

(any

stag

e)or

lung

canc

er(a

nysu

btyp

eor

stag

e),

atte

ndin

gcl

inic

for

the

first

time;

RR

,89

%;

AR

,24

%

2%S

RG

;25

%C

T;40

%R

T;15

%H

T;38

%S

C

391

(I);

378

(I);

365

(C)

NR

Long

itudi

nalt

hree

-ar

mR

CT

Inte

rven

tion

(ful

lscr

eeni

ng):

Com

plet

ion

ofin

terv

entio

nP

RO

Ms,

sum

mar

ies

avai

labl

eto

patie

nt,

and

plac

edon

the

elec

tron

icm

edic

alre

cord

.In

terv

entio

n(t

riage

):S

ame

asfu

llsc

reen

ing,

plus

patie

nts

wer

ein

vite

dto

spea

kto

am

embe

rof

the

psyc

hoso

cial

team

;tr

iage

and

refe

rral

optio

nsav

aila

ble

toth

ose

requ

estin

gan

appo

intm

ent.

Con

trol

(min

imal

scre

enin

g):

Com

plet

ion

ofin

terv

entio

nP

RO

M,

but

nosu

mm

arie

sav

aila

ble

topa

tient

sor

plac

edon

the

med

ical

reco

rd.

Pat

ient

outc

omes

;pr

oces

ses

ofca

re;

heal

thse

rvic

eou

tcom

es

Ele

ctro

nic

inte

ract

ive

tool

;te

leph

one

orem

ailf

/uas

sess

men

t

Long

term

(f/u

asse

ssm

ent

at3

m)

Yes

(con

tinue

don

follo

win

gpa

ge)




Tabl

e1.

Sum

mar

ies

ofth

eM

etho

dolo

gic

Cha

ract

eris

tics

ofth

e24

Stu

dies

(26

artic

les)

Rep

ortin

gon

the

Use

ofP

RO

Ms

As

Inte

rven

tions

inP

atie

nts

With

Can

cer

Rec

eivi

ngA

ctiv

eA

ntic

ance

rTr

eatm

ent

(con

tinue

d)

Aut

hor

and

Yea

rof

Stu

dyS

ettin

g/Lo

catio

nP

atie

ntP

opul

atio

nTy

peof

Trea

tmen

t�N

o.of

Pat

ient

s†N

o.of

HP

sS

tudy

Des

ign

Inte

rven

tion/

Con

trol

Out

com

esA

sses

sed

Met

hod

ofA

dmin

istr

atio

nof

PR

OM

Eva

luat

ion

ofE

ffec

ts

Pat

ient

Rec

eive

dP

RO

MFe

edba

ck

Cle

elan

det

al,40

2011

Hom

e,U

SM

ixed

diag

nose

s(s

tage

ofdi

seas

e)sc

hedu

led

for

thor

acic

surg

ery

for

prim

ary

lung

canc

eror

lung

met

asta

ses;

RR

,N

R;

AR

,21

%

100%

SR

G50

(I);

50(C

)N

RTw

o-ar

mR

CT

Inte

rven

tion:

Com

plet

ion

ofin

terv

entio

nP

RO

Mat

hom

etw

ice

aw

eek

thro

ugh

ate

leph

one-

base

din

tera

ctiv

evo

ice

resp

onse

syst

em;

sym

ptom

info

rmat

ion

inth

efo

rmof

e-m

aila

lert

avai

labl

eto

adva

nced

prac

tice

nurs

eif

sym

ptom

sm

etor

exce

eded

pres

etse

verit

yal

erts

.C

ontr

ol:

Com

plet

ion

ofsa

me

inte

rven

tion

PR

OM

atho

me,

but

nofe

edba

ckav

aila

ble

tocl

inic

ians

.

Pat

ient

outc

omes

;pr

oces

ses

ofca

re

Ele

ctro

nic

inte

ract

ive

tool

atho

me;

pape

r-ba

sed

tool

incl

inic

Inte

rmed

iate

(sev

enf/

utim

epo

ints

with

in4-

6w

)

No

Det

mar

etal

,21

2002

Out

patie

ntcl

inic

,th

eN

ethe

rland

s

Mix

eddi

agno

ses

ofad

vanc

edca

ncer

(typ

ean

dtim

esi

nce

diag

nosi

s);

havi

ngre

ceiv

edat

leas

ttw

ocy

cles

ofpa

lliat

ive

CT;

RR

,71

%;

AR

,22

%

100%

CT

114

(I)‡;

100

(C)

10Lo

ngitu

dina

lcr

osso

ver

two-

arm

RC

T

Inte

rven

tion:

Com

plet

ion

ofin

terv

entio

nP

RO

Man

dsu

mm

arie

sav

aila

ble

tobo

thpa

tient

san

dph

ysic

ians

durin

gco

nsul

tatio

n.C

ontr

ol:

Usu

alca

re.

Pat

ient

outc

omes

;pr

oces

ses

ofca

re

Pap

erto

olin

clin

icLo

ngte

rm(t

hree

f/u

visi

tsaf

ter

base

line)

Yes

Gan

zet

al,22

2000

Out

patie

ntcl

inic

,U

SB

reas

tca

ncer

stag

eI

orII

diag

nose

dbe

twee

n8

man

d5

yea

rlier

;af

ter

com

plet

ion

ofad

juva

ntC

Tor

RT;

RR

,77

%;

AR

,5%

56%

HT

37(I)

;39

(C)

NR

Long

itudi

nalt

wo-

arm

RC

TIn

terv

entio

n:D

aily

com

plet

ion

ofin

terv

entio

nP

RO

Mfo

r28

dbe

fore

base

line;

revi

ewof

info

rmat

ion

and

rece

ipt

ofin

divi

dual

ized

inte

rven

tion

for

thre

esy

mpt

oms:

hot

flash

es,

vagi

nald

ryne

ss,

and

urin

ary

inco

ntin

ence

;f/

uas

sess

men

t.C

ontr

ol:

Dai

lyco

mpl

etio

nof

inte

rven

tion

PR

OM

for

28d

befo

reba

selin

e;in

terv

entio

nw

aspr

ovid

edaf

ter

f/u.

Pat

ient

outc

omes

;he

alth

serv

ice

use

Take

hom

epa

per

tool

/log

book

;pa

per

tool

incl

inic

Long

term

(4-m

f/u

visi

t)Y

es

Girg

iset

al,23

2009

Hom

e,A

ustr

alia

Non

loca

lized

brea

stor

colo

rect

alca

ncer

with

in6

mof

initi

aldi

agno

sis;

RR

,32

%;

AR

,6%

53%

CT;

13%

RT;

2%S

RG

;31

%ot

her

ATR

120

(I);

119

(I);

117

(C)

122§

Long

itudi

nalt

hree

-ar

mR

CT

Inte

rven

tion

(TC

W):

CA

TIus

ing

inte

rven

tion

PR

OM

san

dsu

mm

arie

sav

aila

ble

toTC

W.

Inte

rven

tion

(O/G

P):

CA

TIus

ing

inte

rven

tion

PR

OM

san

dsu

mm

arie

sav

aila

ble

toO

/GP

.C

ontr

ol:

Usu

alca

re.

CA

TIbu

tno

sum

mar

ies

prov

ided

toH

Ps.

Pat

ient

outc

omes

;pr

oces

ses

ofca

re

Tele

phon

eba

sed

Long

term

(3an

d6

maf

ter

base

line)

No

(con

tinue

don

follo

win

gpa

ge)

Kotronoulas et al



Tabl

e1.

Sum

mar

ies

ofth

eM

etho

dolo

gic

Cha

ract

eris

tics

ofth

e24

Stu

dies

(26

artic

les)

Rep

ortin

gon

the

Use

ofP

RO

Ms

As

Inte

rven

tions

inP

atie

nts

With

Can

cer

Rec

eivi

ngA

ctiv

eA

ntic

ance

rTr

eatm

ent

(con

tinue

d)

Aut

hor

and

Yea

rof

Stu

dyS

ettin

g/Lo

catio

nP

atie

ntP

opul

atio

nTy

peof

Trea

tmen

t�N

o.of

Pat

ient

s†N

o.of

HP

sS

tudy

Des

ign

Inte

rven

tion/

Con

trol

Out

com

esA

sses

sed

Met

hod

ofA

dmin

istr

atio

nof

PR

OM

Eva

luat

ion

ofE

ffec

ts

Pat

ient

Rec

eive

dP

RO

MFe

edba

ck

Hila

rius

etal

,24

2008

Out

patie

ntcl

inic

,th

eN

ethe

rland

s

Mix

edca

ncer

diag

nose

s(t

ype

and

stag

e)at

the

star

tof

CT

trea

tmen

t;R

R,

83%

;A

R,

26.5

%

100%

CT

148

(I);

150

(C)

10Lo

ngitu

dina

lse

quen

tialt

wo-

arm

coho

rt

Inte

rven

tion:

Com

plet

ion

ofin

terv

entio

nP

RO

Man

dsu

mm

arie

sav

aila

ble

tobo

thpa

tient

san

dnu

rses

durin

gco

nsul

tatio

n.C

ontr

ol:

Com

plet

ion

ofin

terv

entio

nP

RO

M,

but

sum

mar

ies

unav

aila

ble

tonu

rses

durin

gco

nsul

tatio

n.

Pat

ient

outc

omes

;pr

oces

ses

ofca

re

Ele

ctro

nic

inte

ract

ive

tool

Long

term

(fou

rf/

uvi

sits

)Y

es

Hoe

kstr

aet

al,25

2006

GP

prac

tice

and

hom

e,th

eN

ethe

rland

s

Adv

ance

dbr

east

,lu

ng,

orG

Ica

ncer

with

alif

eex

pect

ancy

of1-

12m

;R

R,

89%

;A

R,

32%

100%

PS

C69

(I)‡;

77(C

)89

Long

itudi

nalt

wo-

arm

clus

tere

dR

CT

Inte

rven

tion:

Wee

kly

self-

asse

ssm

ent

ofph

ysic

alsy

mpt

oms

atho

me

thro

ugh

use

ofth

ein

terv

entio

nP

RO

M.

Con

trol

:S

tand

ard

care

.

Pat

ient

outc

omes

Take

-hom

epa

per

tool

/log

book

Long

term

(eve

ryot

her

mon

th)

NR

Kea

rney

etal

,41

2009

Hom

ean

dou

tpat

ient

clin

ic,

UK

Bre

ast,

lung

,or

colo

rect

alca

ncer

(any

stag

e)at

the

initi

atio

nof

ane

wco

urse

ofC

Ttr

eatm

ent

(any

CT

line)

;R

R,

NR

;A

R,

23%

100%

CT

56(I)

;56

(C)

NR

Long

itudi

nalt

wo-

arm

RC

TIn

terv

entio

n:C

ompl

etio

nof

inte

rven

tion

PR

OM

onm

obile

phon

eat

hom

eon

days

1-14

post

-CT

adm

inis

trat

ion;

sym

ptom

info

rmat

ion

avai

labl

eto

clin

icia

nsin

real

-tim

ein

the

form

ofal

erts

(am

ber:

mild

/mod

erat

ese

verit

y;re

d:se

vere

orlif

e-th

reat

enin

g);

clin

icia

nsco

ntac

ted

the

patie

ntw

ithin

48-7

2h

(am

ber)

or1

h(r

ed).

Con

trol

:S

tand

ard

care

(writ

ten

and

verb

alin

form

atio

n).

Pat

ient

outc

omes

Ele

ctro

nic

inte

ract

ive

tool

atho

me;

pape

r-ba

sed

tool

incl

inic

Long

term

(fou

rf/

utim

epo

ints

with

in12

-16

w)

Yes

Klin

kham

mer

-S

chal

keet

al,26

2012

Inpa

tient

surg

ery

clin

ics,

Ger

man

y

New

lydi

agno

sed

brea

stca

ncer

(any

stag

e)at

disc

harg

eaf

ter

initi

alsu

rgic

altr

eatm

ent;

RR

,82

%;

AR

,15

%

100%

SR

G99

(I);

100

(C)

146

Long

itudi

nalt

wo-

arm

RC

TIn

terv

entio

n:C

ompl

etio

nof

inte

rven

tion

PR

OM

bypa

tient

and

heal

thst

atus

form

byph

ysic

ian;

profi

les

avai

labl

eto

expe

rts;

expe

rtop

inio

nav

aila

ble

toco

ordi

natin

gpr

actit

ione

rsw

hoar

rang

edQ

oLth

erap

yco

nsis

ting

ofup

tofiv

est

anda

rdiz

edtr

eatm

ents

.C

ontr

ol:

Com

plet

ion

ofin

terv

entio

nP

RO

M,

but

profi

les

and

expe

rtop

inio

nsun

avai

labl

eto

prac

titio

ners

.

Pat

ient

outc

omes

;pr

oces

ses

ofca

re

Ele

ctro

nic

inte

ract

ive

tool

Long

term

(f/u

asse

ssm

ents

at3,

6,9,

and

12m

afte

rba

selin

e)

No

(con

tinue

don

follo

win

gpa

ge)




Tabl

e1.

Sum

mar

ies

ofth

eM

etho

dolo

gic

Cha

ract

eris

tics

ofth

e24

Stu

dies

(26

artic

les)

Rep

ortin

gon

the

Use

ofP

RO

Ms

As

Inte

rven

tions

inP

atie

nts

With

Can

cer

Rec

eivi

ngA

ctiv

eA

ntic

ance

rTr

eatm

ent

(con

tinue

d)

Aut

hor

and

Yea

rof

Stu

dyS

ettin

g/Lo

catio

nP

atie

ntP

opul

atio

nTy

peof

Trea

tmen

t�N

o.of

Pat

ient

s†N

o.of

HP

sS

tudy

Des

ign

Inte

rven

tion/

Con

trol

Out

com

esA

sses

sed

Met

hod

ofA

dmin

istr

atio

nof

PR

OM

Eva

luat

ion

ofE

ffec

ts

Pat

ient

Rec

eive

dP

RO

MFe

edba

ck

Kor

nblit

het

al,42

2006

Hom

e,U

SB

reas

t,co

lon,

orpr

osta

teca

ncer

(sta

ges

IIIor

IV)

with

inth

efir

st2

mof

activ

etr

eatm

ent;

life

expe

ctan

cyof

�12

m;

RR

,82

%;

AR

,62

%

100%

ATR

96(I)

;93

(C)

NR

Long

itudi

nalt

wo-

arm

RC

TIn

terv

entio

n:C

ompl

etio

nof

inte

rven

tion

PR

OM

sat

hom

em

onth

lyfo

r6

mth

roug

hTM

inad

ditio

nto

EM

;fe

edba

ckav

aila

ble

toon

colo

gynu

rse

ifle

vels

ofdi

stre

ssab

ove

pres

etcu

t-of

fsc

ores

.In

divi

dual

ized

disc

ussi

onan

dtr

eatm

ent

reco

mm

enda

tion

durin

gf/

uca

lls.

Con

trol

:S

tand

ard

care

and

EM

only

.

Pat

ient

outc

omes

;pr

oces

ses

ofca

re;

heal

thse

rvic

esou

tcom

es

Tele

phon

eba

sed

Long

term

(tw

of/

uas

sess

men

tsat

6an

d9

m)

NR

Mau

nsel

let

al,27

1996

Inpa

tient

clin

ic,

Can

ada

New

lydi

agno

sed

brea

stca

ncer

(any

stag

e)af

ter

initi

alsu

rgic

altr

eatm

ent;

RR

,89

%;

AR

,10

%

67%

RT;

30%

CT;

46%

HT

130

(I);

131

(C)

NR

Long

itudi

nalt

wo-

arm

RC

TIn

terv

entio

n:B

rief

psyc

hoso

cial

inte

rven

tion

byso

cial

wor

ker

post

surg

ery

and

f/u

scre

enin

gfo

rps

ycho

logi

cald

istr

ess

with

inte

rven

tion

PR

OM

;fu

rthe

rin

terv

entio

nfo

rhi

ghly

dist

ress

edpa

tient

s.C

ontr

ol:

Brie

fps

ycho

soci

alin

terv

entio

nby

soci

alw

orke

rpo

stsu

rger

ybu

tno

f/u

scre

enin

g.

Pat

ient

outc

omes

;pr

oces

ses

ofca

re;

heal

thse

rvic

eus

e

Tele

phon

eba

sed

Long

term

(3an

d12

m)

NR

McL

achl

anet

al,28

2001

Out

patie

ntcl

inic

,A

ustr

alia

Mix

edca

ncer

diag

nose

s(t

ype,

stag

e,an

dtim

esi

nce

diag

nosi

s);

havi

ngat

tend

edat

leas

ton

eco

nsul

tatio

n;R

R,

59%

;A

R,

29%

26%

SC

;32

%C

T�

RT;

5%ot

her

ATR

296

(I);

154

(C)

NR

Long

itudi

nalt

wo-

arm

RC

TIn

terv

entio

n:A

sses

smen

tw

ithin

terv

entio

n,P

RO

Mbe

fore

cons

ulta

tion,

and

sum

mar

yim

med

iate

lyav

aila

ble

toco

nsul

tant

s.In

divi

dual

ized

man

agem

ent

plan

base

don

patie

nt’s

resp

onse

s.C

ontr

ol:

Con

vent

iona

lcl

inic

alen

coun

ter

and

self-

repo

rted

info

rmat

ion

unav

aila

ble

toco

nsul

tant

s.

Pat

ient

outc

omes

;pr

oces

ses

ofca

re

Ele

ctro

nic

inte

ract

ive

tool

Long

term

(2an

d6

maf

ter

base

line)

NR

Mill

set

al,29

2009

Inpa

tient

clin

ic,

UK

Inop

erab

lelu

ngca

ncer

,an

ysu

btyp

e;R

R,

51%

;A

R,

50%

61%

CT;

17%

RT;

16%

CT

plus

RT;

6%S

C

57(I)

;58

(C)

NR

Long

itudi

nalt

wo-

arm

RC

TIn

terv

entio

n:W

eekl

yco

mpl

etio

nof

inte

rven

tion

PR

OM

atho

me;

patie

nts

wer

eas

ked

tosh

are

info

rmat

ion

with

any

HP

invo

lved

inth

eir

care

.C

ontr

ol:

Usu

alca

re.

Pat

ient

outc

omes

;pr

oces

ses

ofca

re

Take

-hom

epa

per

tool

Long

term

(16

w)

NA

(con

tinue

don

follo

win

gpa

ge)

Kotronoulas et al



Tabl

e1.

Sum

mar

ies

ofth

eM

etho

dolo

gic

Cha

ract

eris

tics

ofth

e24

Stu

dies

(26

artic

les)

Rep

ortin

gon

the

Use

ofP

RO

Ms

As

Inte

rven

tions

inP

atie

nts

With

Can

cer

Rec

eivi

ngA

ctiv

eA

ntic

ance

rTr

eatm

ent

(con

tinue

d)

Aut

hor

and

Yea

rof

Stu

dyS

ettin

g/Lo

catio

nP

atie

ntP

opul

atio

nTy

peof

Trea

tmen

t�N

o.of

Pat

ient

s†N

o.of

HP

sS

tudy

Des

ign

Inte

rven

tion/

Con

trol

Out

com

esA

sses

sed

Met

hod

ofA

dmin

istr

atio

nof

PR

OM

Eva

luat

ion

ofE

ffec

ts

Pat

ient

Rec

eive

dP

RO

MFe

edba

ck

Nic

klas

son

etal

,30

2013

Out

patie

ntcl

inic

,S

wed

enIn

cura

ble

lung

canc

er(a

nysu

btyp

eor

stag

e)or

mes

othe

liom

aw

itha

life

expe

ctan

cyat

the

first

clin

icvi

sit

of�

3m

;R

R,

75%

;A

R,

1%

78%

CT;

42%

RT;

9%S

C85

(I);

88(C

)22

Long

itudi

nalt

wo-

arm

RC

TIn

terv

entio

n:C

ompl

etio

nof

com

pute

rized

inte

rven

tion

PR

OM

befo

reco

nsul

tatio

n;su

mm

arie

sw

ere

avai

labl

eto

cons

ultin

gph

ysic

ians

.C

ontr

ol:

Com

plet

ion

ofpa

per-

and-

penc

ilin

terv

entio

nP

RO

Mbe

fore

cons

ulta

tion,

but

sum

mar

ies

wer

eun

avai

labl

eto

cons

ultin

gph

ysic

ians

.

Pro

cess

esof

care

;he

alth

serv

ice

outc

omes

Ele

ctro

nic

inte

ract

ive

tool

;pa

per

tool

incl

inic

Long

term

(8-1

2w

off/

uvi

sits

)N

o

Ros

enbl

oom

etal

,31

2007

Out

patie

ntcl

inic

,U

SA

dvan

ced

brea

st,

lung

,or

colo

rect

alca

ncer

with

alif

eex

pect

ancy

of�

6m

durin

gC

Ttr

eatm

ent;

RR

,N

R,

and

AR

:28

%

100%

CT

69(I)

;71

(C);

73(C

)N

RLo

ngitu

dina

lthr

ee-

arm

RC

TIn

terv

entio

n:A

sses

smen

tw

ithin

terv

entio

nP

RO

Mfo

llow

edby

stru

ctur

edin

terv

iew

with

trea

ting

nurs

eab

out

patie

nt’s

resp

onse

s.A

sses

smen

tco

ntro

l:A

sses

smen

tw

ithin

terv

entio

nP

RO

Mfo

llow

edby

feed

back

totr

eatin

gnu

rses

,bu

tno

inte

rvie

w.

Full

cont

rol:

Ass

essm

ent

with

outc

ome

PR

OM

,bu

tno

inte

rvie

ww

ithor

feed

back

totr

eatin

gnu

rses

.

Pat

ient

outc

omes

;pr

oces

ses

ofca

re

Pap

erto

olin

clin

icLo

ngte

rm(f

our

f/u

visi

ts)

No

Rul

and

etal

,32

2010

Inpa

tient

and

outp

atie

ntcl

inic

s,N

orw

ay

New

lydi

agno

sed

orre

curr

ent

hem

atol

ogic

mal

igna

ncy

atth

est

art

oftr

eatm

ent;

RR

,90

%;

AR

,19

%

68%

CT;

34%

SC

T75

(I);

70(C

)N

RLo

ngitu

dina

ltw

o-ar

mR

CT

Inte

rven

tion:

Inte

rven

tion

PR

OM

adm

inis

tere

ddu

ring

inpa

tient

,ou

tpat

ient

,an

dal

lf/u

visi

ts.

Ass

essm

ent

sum

mar

ies

avai

labl

eto

HP

s.C

ontr

ol:

Inte

rven

tion

PR

OM

adm

inis

tere

ddu

ring

inpa

tient

,ou

tpat

ient

,an

dal

lf/u

visi

ts.

Ass

essm

ent

sum

mar

ies

not

avai

labl

eto

HP

s.

Pat

ient

outc

omes

;pr

oces

ses

ofca

re

Ele

ctro

nic

inte

ract

ive

tool

Long

term

(�fo

urfo

llow

-up

visi

ts)

NR

Sar

na,33 19

98O

utpa

tient

clin

ics,

US

Adv

ance

dlu

ngca

ncer

(any

subt

ype)

;ne

wly

diag

nose

d;R

R,

83%

;A

R,

56%

88%

CT;

23%

RT

48¶

NR

Long

itudi

nalt

wo-

arm

RC

TIn

terv

entio

n:C

ompl

etio

nof

inte

rven

tion

PR

OM

and

sum

mar

ies

avai

labl

eto

staf

fnu

rses

for

disc

ussi

onw

ithth

epa

tient

.C

ontr

ol:

Com

plet

ion

ofin

terv

entio

nP

RO

M,

but

sum

mar

ies

unav

aila

ble

tost

aff

nurs

es.

Pat

ient

outc

omes

Pap

erto

olin

clin

icLo

ngte

rm(s

ixas

sess

men

tpo

ints

with

ina

6-m

perio

d)

No

(con

tinue

don

follo

win

gpa

ge)




Tabl

e1.

Sum

mar

ies

ofth

eM

etho

dolo

gic

Cha

ract

eris

tics

ofth

e24

Stu

dies

(26

artic

les)

Rep

ortin

gon

the

Use

ofP

RO

Ms

As

Inte

rven

tions

inP

atie

nts

With

Can

cer

Rec

eivi

ngA

ctiv

eA

ntic

ance

rTr

eatm

ent

(con

tinue

d)

Aut

hor

and

Yea

rof

Stu

dyS

ettin

g/Lo

catio

nP

atie

ntP

opul

atio

nTy

peof

Trea

tmen

t�N

o.of

Pat

ient

s†N

o.of

HP

sS

tudy

Des

ign

Inte

rven

tion/

Con

trol

Out

com

esA

sses

sed

Met

hod

ofA

dmin

istr

atio

nof

PR

OM

Eva

luat

ion

ofE

ffec

ts

Pat

ient

Rec

eive

dP

RO

MFe

edba

ck

Taen

zer

etal

,34

2000

Out

patie

ntcl

inic

,C

anad

aP

rimar

y,se

cond

ary,

orm

etas

tatic

lung

canc

erof

any

stag

e;an

aver

age

of51

mpo

stdi

agno

sis;

RR

,70

%;

AR

,N

R

NR

%S

C;

NR

%A

TR27

(I);

26(C

)N

RS

eque

ntia

lpre

-an

dpo

stsc

reen

,tw

o-ar

mco

hort

Inte

rven

tion:

Com

plet

ion

ofin

terv

entio

nP

RO

Mbe

fore

cons

ulta

tion

and

sum

mar

ies

prov

ided

toH

Ps.

Con

trol

:U

sual

care

.

Pro

cess

esof

care

Ele

ctro

nic

inte

ract

ive

tool

(I);

pape

rto

olin

clin

ic(C

)

Sho

rtte

rm(s

ame

day

asco

nsul

tatio

n)

No

Take

uchi

etal

,35

2011

�

Out

patie

ntcl

inic

,U

KM

ixed

canc

erdi

agno

ses

(typ

ean

dst

age)

atth

est

art

oftr

eatm

ent;

RR

,65

%;

AR

,37

%

100%

ATR

100

(I);

46(C

);52

(C)

28Lo

ngitu

dina

lthr

ee-

arm

RC

TIn

terv

entio

n:C

ompl

etio

nof

touc

h-sc

reen

inte

rven

tion

PR

OM

sbe

fore

clin

icvi

sit

and

feed

back

avai

labl

eto

phys

icia

ns.

Att

entio

n-co

ntro

l:C

ompl

etio

nof

inte

rven

tion

PR

OM

sbe

fore

clin

icvi

sit,

but

feed

back

unav

aila

ble

toph

ysic

ians

.C

ontr

ol:

Sta

ndar

dca

re.

Pro

cess

esof

care

Ele

ctro

nic

inte

ract

ive

tool

Long

term

(fou

rtim

epo

ints

with

in6

m)

No

Thew

eset

al,36

2009

Rur

alou

tpat

ient

clin

ics;

hom

e;A

ustr

alia

Mix

edca

ncer

diag

nose

s(t

ype

and

stag

e);

new

lydi

agno

sed

atth

efir

stcl

inic

visi

t;R

R,

81%

;A

R,

37%

76%

SR

G;

66%

CT;

53%

RT;

33%

HT

43(I)

;40

(C)

NR

Seq

uent

ialp

re-

and

post

scre

en,

two-

arm

coho

rt

Inte

rven

tion:

Com

plet

ion

ofin

terv

entio

nP

RO

Man

dfe

edba

ckto

nurs

ing

staf

f;pa

tient

asse

ssm

ent

ofpr

oble

ms

and

conc

erns

ifsc

ore

abov

ecu

toff

scor

e.C

ontr

ol:

Usu

alca

re;

noin

terv

entio

nP

RO

Mad

min

iste

red.

Pat

ient

outc

omes

;pr

oces

ses

ofca

re

Pap

erto

olin

clin

ic;

mai

led

f/u

asse

ssm

ents

Long

term

(one

f/u

at6

maf

ter

inte

rven

tion)

NR

Trow

brid

geet

al,37

1997

Out

patie

ntcl

inic

,U

SM

ixed

diag

nose

sof

recu

rren

tor

met

asta

ticso

lidor

hem

atol

ogic

canc

ers

orsa

rcom

as;

RR

,N

R;

AR

,N

R

100%

ATR

160

(I);

160

(C)

13Tw

o-ar

mR

CT

Inte

rven

tion:

Com

plet

ion

ofin

terv

entio

nP

RO

Mbe

fore

cons

ulta

tion

and

sum

mar

ies

prov

ided

toco

nsul

tant

;di

scus

sion

ofse

lf-re

port

edin

form

atio

n.C

ontr

ol:

Com

plet

ion

ofin

terv

entio

nP

RO

Mbe

fore

cons

ulta

tion,

but

sum

mar

ies

unav

aila

ble

toco

nsul

tant

.

Pat

ient

outc

omes

;pr

oces

ses

ofca

re

Pap

erto

olin

clin

ic;

mai

led

asse

ssm

ents

Inte

rmed

iate

(4w

afte

rin

terv

entio

n)

No

Vel

ikov

aet

al,38

2004

�

Out

patie

ntcl

inic

,U

KM

ixed

canc

erdi

agno

ses

(typ

ean

dst

age)

atth

est

art

oftr

eatm

ent;

RR

,65

%;

AR

,37

%

76%

CT;

21%

BT;

2%H

T;1%

f/u

144

(I);

70(C

);72

(C)

28Lo

ngitu

dina

lthr

ee-

arm

RC

TIn

terv

entio

n:C

ompl

etio

nof

touc

h-sc

reen

inte

rven

tion

PR

OM

sbe

fore

clin

icvi

sit

and

feed

back

avai

labl

eto

phys

icia

ns.

Att

entio

n-co

ntro

l:C

ompl

etio

nof

inte

rven

tion

PR

OM

sbe

fore

clin

icvi

sit,

but

feed

back

unav

aila

ble

toph

ysic

ians

.C

ontr

ol:

Sta

ndar

dca

re.

Pat

ient

outc

omes

;pr

oces

ses

ofca

re

Ele

ctro

nic

inte

ract

ive

tool

Long

term

(fou

rtim

epo

ints

with

in6

m)

No

(con

tinue

don

follo

win

gpa

ge)

Kotronoulas et al



Tabl

e1.

Sum

mar

ies

ofth

eM

etho

dolo

gic

Cha

ract

eris

tics

ofth

e24

Stu

dies

(26

artic

les)

Rep

ortin

gon

the

Use

ofP

RO

Ms

As

Inte

rven

tions

inP

atie

nts

With

Can

cer

Rec

eivi

ngA

ctiv

eA

ntic

ance

rTr

eatm

ent

(con

tinue

d)

Aut

hor

and

Yea

rof

Stu

dyS

ettin

g/Lo

catio

nP

atie

ntP

opul

atio

nTy

peof

Trea

tmen

t�N

o.of

Pat

ient

s†N

o.of

HP

sS

tudy

Des

ign

Inte

rven

tion/

Con

trol

Out

com

esA

sses

sed

Met

hod

ofA

dmin

istr

atio

nof

PR

OM

Eva

luat

ion

ofE

ffec

ts

Pat

ient

Rec

eive

dP

RO

MFe

edba

ck

Vel

ikov

aet

al,39

2010

�

Out

patie

ntcl

inic

,U

KM

ixed

canc

erdi

agno

ses

(typ

ean

dst

age)

atth

est

art

oftr

eatm

ent;

RR

,65

%;

AR

,37

%

76%

CT;

21%

BT;

2%H

T;1%

f/u

144

(I);

70(C

);72

(C)

28Lo

ngitu

dina

lthr

ee-

arm

RC

TIn

terv

entio

n:C

ompl

etio

nof

touc

h-sc

reen

inte

rven

tion

PR

OM

sbe

fore

clin

icvi

sit

and

feed

back

ofre

sults

avai

labl

eto

phys

icia

ns.

Att

entio

n-co

ntro

l:C

ompl

etio

nof

inte

rven

tion

PR

OM

sbe

fore

clin

icvi

sit,

but

feed

back

unav

aila

ble

toph

ysic

ians

.C

ontr

ol:

Sta

ndar

dca

re.

Pro

cess

esof

care

Ele

ctro

nic

inte

ract

ive

tool

Long

term

(fou

rtim

epo

ints

with

in6

m)

No

Abb

revi

atio

ns:A

R,a

ttrit

ion

rate

;ATR

,act

ive

trea

tmen

t;B

T,bi

olog

ical

ther

apy;

C,c

ontr

ol;C

ATI

,com

pute

r-as

sist

edte

leph

one

inte

rvie

w;C

T,ch

emot

hera

py;d

,day

s;E

M,e

duca

tiona

lmat

eria

ls;E

SR

A-C

,Ele

ctro

nic

Sel

f-R

epor

tAss

essm

ent-

Can

cer;

f/u,

follo

w-u

p;G

P,g

ener

alpr

actit

ione

r;h,

hour

s;H

P,h

ealth

care

prof

essi

onal

;HT,

horm

onal

ther

apy;

I,in

terv

entio

n;m

,mon

ths;

MD

,med

ical

;NA

,not

appl

icab

le;N

R,n

otre

port

ed;

O,

onco

logi

st;

PR

OM

,pa

tient

-rep

orte

dou

tcom

em

easu

re;

PS

C,

palli

ativ

esu

ppor

tive

care

;Q

oL,

qual

ityof

life;

RC

T,ra

ndom

ized

cont

rolle

dtr

ial;

RR

,re

spon

sera

te;

RT,

radi

othe

rapy

;S

C,

supp

ortiv

eca

re;

SC

T,st

em-c

ellt

rans

plan

tatio

n;S

RG

,su

rger

y;TC

W,

tele

phon

eca

sew

orke

r;TM

,te

leph

one

mon

itorin

g;U

K,

Uni

ted

Kin

gdom

;U

S,

Uni

ted

Sta

tes;

w,

wee

ks;

y,ye

ars.

�P

erce

ntag

esva

lidfo

rto

tals

ampl

e.†S

ampl

esi

zes

ofpa

tient

sas

rand

omly

assi

gned

(RC

Ts)

and

cons

ente

d(n

on-R

CTs

)at

base

line.

‡Phy

sici

ans,

rath

erth

anpa

tient

s,w

ere

rand

omly

assi

gned

.§E

stim

ated

asth

eto

talo

f3

TCW

san

d11

9O

/GP

s.¶

Gro

upsi

zes

wer

eno

tre

port

ed.

�Art

icle

sar

eba

sed

onda

tafr

omth

esa

me

stud

y;di

ffer

ent

sam

ple

size

san

dou

tcom

esar

eev

alua

ted

inea

char

ticle

.




Tabl

e2.

Mai

nFi

ndin

gsan

dA

sses

smen

tof

Ris

kof

Bia

sin

the

20R

CTs

and

Four

Non

-RC

TsId

entifi

edfo

rTh

isR

evie

w

Aut

hor

and

Yea

r

Mai

nS

tudy

Find

ings

�

Gui

delin

esW

ere

Use

dto

Gui

deC

linic

ian

Res

pons

e

Ris

kof

Bia

s†

Sel

ectio

nB

ias

Per

form

ance

Bia

s:B

lindi

ngP

artic

ipan

tsan

dP

erso

nnel

Det

ectio

nB

ias:

Blin

ding

ofO

utco

me

Ass

essm

ent

Att

ritio

nB

ias:

Inco

mpl

ete

Out

com

eD

ata

Rep

ortin

gB

ias:

Sel

ectiv

eO

utco

me

Rep

ortin

g

Oth

erS

ourc

esof

Bia

sP

atie

ntO

utco

mes

Pro

cess

esof

Car

eH

ealth

Ser

vice

Out

com

es

Ran

dom

Seq

uenc

eG

ener

atio

nA

lloca

tion

Con

ceal

men

t

RC

Ts Ber

ryet

al,17

2011

Inte

rven

tion

effe

cts

depe

nded

onw

heth

era

sym

ptom

/QoL

issu

ew

asre

port

edat

thre

shol

d(P

�.0

3).

Whe

nre

port

edat

thre

shol

d,th

ein

terv

entio

nre

sulte

din

a29

%in

crea

sein

the

odds

ofth

eis

sue

bein

gdi

scus

sed

com

pare

dw

ithth

eC

G.

This

was

evid

ent

for

conc

entr

atio

n,co

gniti

vefu

nctio

n,im

pact

onse

xual

activ

ities

and

inte

rest

,an

dso

cial

func

tion.

No

EG

/CG

diff

eren

ces

(P�

.35)

for

the

aver

age

leng

thof

clin

icvi

sits

.C

linic

ians

agre

edth

atth

ein

terv

entio

nw

asus

eful

inid

entif

ying

appr

opria

tesy

mpt

om/Q

oLis

sues

(67.

8%),

guid

ing

the

inte

rvie

w(6

4.3%

),pr

omot

ing

com

mun

icat

ion

(50%

),an

did

entif

ying

appr

opria

tear

eas

for

refe

rral

(53.

6%).

—N

oLo

wLo

wH

igh

Unc

lear

Low

Hig

hH

igh

Bra

eken

etal

,19

2011

—N

osi

gnifi

cant

inte

rven

tion

effe

cts

wer

eob

serv

edfo

rth

eto

talN

o.of

patie

nts

refe

rred

tops

ycho

soci

alca

repr

ovid

ers

at3

(P�

.32)

,9

(P�

.22)

,or

12m

(P�

.44)

.M

ore

patie

nts

inth

eE

Gbr

ough

tup

thei

rne

edfo

rps

ycho

soci

alca

redu

ring

cons

ulta

tion

(P�

.04)

.E

Gw

ere

refe

rred

toso

cial

wor

kers

atan

earli

erst

age

than

CG

(P�

.01)

.N

osi

gnifi

cant

inte

rven

tion

effe

cton

impr

ovin

gpa

tient

-clin

icia

nco

mm

unic

atio

nab

out

psyc

hoso

cial

prob

lem

s;no

effe

cton

patie

nts’

satis

fact

ion

with

com

mun

icat

ion

with

clin

icia

ns.

—N

oU

ncle

arU

ncle

arH

igh

Hig

hH

igh

Hig

hLo

w

Car

lson

etal

,20

2010

Onl

ya

mar

gina

llysi

gnifi

cant

mai

nef

fect

ofst

udy

cond

ition

atfo

llow

-up

for

dist

ress

scor

es(P

�.0

9).

Sig

nific

antly

few

erpa

tient

sin

the

tria

gegr

oup

(36%

)ex

ceed

edth

edi

stre

sscu

tof

fv

46%

and

48.7

%in

full

scre

enin

gan

dm

inim

alsc

reen

ing

grou

ps,

resp

ectiv

ely

(P�

.005

).N

oE

G/C

Gdi

ffer

ence

sin

anxi

ety

orde

pres

sion

scor

esat

3m

over

allo

rw

ithin

eith

erth

elu

ngor

brea

stgr

oups

.

No

diff

eren

ces

betw

een

stud

yco

nditi

ons

inre

ferr

als

mad

eto

psyc

hoso

cial

care

(P�

.05)

befo

reor

afte

rfo

llow

-up

.R

ecei

ving

are

ferr

alw

aslin

ked

tole

ssim

prov

emen

ton

the

dist

ress

scor

e.

No

diff

eren

ces

betw

een

full

scre

enin

gan

dm

inim

alsc

reen

ing

inpa

tient

self-

refe

rral

s(1

4.3%

v10

.3%

).

Yes

Low

Low

Hig

hH

igh

Hig

hU

ncle

arLo

w

(con

tinue

don

follo

win

gpa

ge)

Kotronoulas et al



Tabl

e2.

Mai

nFi

ndin

gsan

dA

sses

smen

tof

Ris

kof

Bia

sin

the

20R

CTs

and

Four

Non

-RC

TsId

entifi

edfo

rTh

isR

evie

w(c

ontin

ued)

Aut

hor

and

Yea

r

Mai

nS

tudy

Find

ings

�

Gui

delin

esW

ere

Use

dto

Gui

deC

linic

ian

Res

pons

e

Ris

kof

Bia

s†

Sel

ectio

nB

ias

Per

form

ance

Bia

s:B

lindi

ngP

artic

ipan

tsan

dP

erso

nnel

Det

ectio

nB

ias:

Blin

ding

ofO

utco

me

Ass

essm

ent

Att

ritio

nB

ias:

Inco

mpl

ete

Out

com

eD

ata

Rep

ortin

gB

ias:

Sel

ectiv

eO

utco

me

Rep

ortin

g

Oth

erS

ourc

esof

Bia

sP

atie

ntO

utco

mes

Pro

cess

esof

Car

eH

ealth

Ser

vice

Out

com

es

Ran

dom

Seq

uenc

eG

ener

atio

nA

lloca

tion

Con

ceal

men

t

Cle

elan

det

al,40

2011

EG

sign

ifica

ntly

grea

ter

redu

ctio

nin

over

alls

ympt

omth

resh

old

even

tsdu

ring

the

4-w

eek

tria

lpe

riod

(19%

v8%

;P

�.0

03).

Sym

ptom

thre

shol

dev

ents

for

pain

,di

stre

ss,

dist

urbe

dsl

eep,

shor

tnes

sof

brea

th,

and

cons

tipat

ion

wer

em

ore

inth

eC

Gat

wee

k4.

No

EG

/C

Gdi

ffer

ence

sin

mea

nsy

mpt

omse

verit

ych

ange

sat

the

end

of4

wee

ks.

Gre

ater

redu

ctio

nin

mea

nsy

mpt

omin

terf

eren

ceov

ertim

ein

EG

(P�

.02)

.

EG

sign

ifica

ntly

mor

eco

mfo

rtab

lew

ithin

terv

entio

n(P

�.0

3)an

dm

ore

likel

yto

rate

the

inte

rven

tion

asea

syto

use

(P�

.01)

com

pare

dw

ithC

G.

Bot

hgr

oups

expr

esse

dsa

tisfa

ctio

nw

ithth

ein

terv

entio

nan

dag

reed

for

itto

beus

edin

rout

ine

clin

ical

prac

tice.

—N

oU

ncle

arLo

wLo

wLo

wH

igh

Unc

lear

Hig

h

Det

mar

etal

,21

2002

No

EG

/CG

diff

eren

ces

atth

efo

urth

visi

tfo

ran

yof

the

QoL

scal

es.

Asi

gnifi

cant

lygr

eate

rpe

rcen

tage

ofpa

tient

sin

the

EG

vth

eC

Gex

hibi

ted

impr

ovem

ent

over

time

inm

enta

lhea

lth(4

3%v

30%

;P

�.0

4)an

dro

lefu

nctio

ning

(22%

v11

%;

P�

.05)

.

Ten

of12

QoL

issu

esw

ere

disc

usse

dm

ore

freq

uent

lyin

the

EG

,es

peci

ally

soci

alfu

nctio

ning

,fa

tigue

,an

ddy

spne

a(P

�.0

5).

No

EG

/C

Gdi

ffer

ence

sin

exac

tor

glob

alph

ysic

ian-

patie

ntag

reem

ent,

orin

mea

nnu

mbe

rof

QoL

-rel

ated

patie

ntm

anag

emen

tac

tions

take

npe

rpa

tient

.P

atie

ntan

dph

ysic

ian

satis

fact

ion

was

high

inbo

thgr

oups

.N

odi

ffer

ence

sin

mea

ndu

ratio

nof

visi

ts.

Inth

eE

G,

the

QoL

sum

mar

ypr

ofile

prov

ided

anac

cura

tepi

ctur

eof

patie

ntfu

nctio

ning

and

wel

l-bei

ng(9

7%),

and

itw

ould

beus

eful

asa

stan

dard

part

ofth

eou

tpat

ient

clin

icpr

oced

ure

(87%

).

—N

oU

ncle

arU

ncle

arU

ncle

arH

igh

Low

Unc

lear

Hig

h

Gan

zet

al,22

2000

Cha

nge

scor

esfo

rm

enop

ausa

lsy

mpt

oms

(P�

.001

)an

dse

xual

func

tioni

ng(P

�.0

2)di

ffer

edsi

gnifi

cant

lybe

twee

ngr

oups

,w

ithE

Gre

port

ing

few

erse

vere

sym

ptom

san

dbe

tter

sexu

alfu

nctio

ning

atfo

llow

-up.

No

EG

/CG

diff

eren

ces

inte

rms

ofvi

talit

y(P

�.7

7).

—W

omen

inbo

thth

eE

Gan

dC

Gso

ught

out

addi

tiona

linf

orm

atio

nab

out

thei

rsy

mpt

oms,

atab

out

the

sam

era

te.

Asi

mila

rpe

rcen

tage

ofw

omen

inea

chgr

oup

rece

ived

som

efo

rmof

psyc

holo

gica

lre

ferr

al.

Wom

enin

the

EG

used

med

icat

ions

mor

efr

eque

ntly

.

Yes

Unc

lear

Unc

lear

Unc

lear

Unc

lear

Low

Unc

lear

Hig

h

(con

tinue

don

follo

win

gpa

ge)




Tabl

e2.

Mai

nFi

ndin

gsan

dA

sses

smen

tof

Ris

kof

Bia

sin

the

20R

CTs

and

Four

Non

-RC

TsId

entifi

edfo

rTh

isR

evie

w(c

ontin

ued)

Aut

hor

and

Yea

r

Mai

nS

tudy

Find

ings

�

Gui

delin

esW

ere

Use

dto

Gui

deC

linic

ian

Res

pons

e

Ris

kof

Bia

s†

Sel

ectio

nB

ias

Per

form

ance

Bia

s:B

lindi

ngP

artic

ipan

tsan

dP

erso

nnel

Det

ectio

nB

ias:

Blin

ding

ofO

utco

me

Ass

essm

ent

Att

ritio

nB

ias:

Inco

mpl

ete

Out

com

eD

ata

Rep

ortin

gB

ias:

Sel

ectiv

eO

utco

me

Rep

ortin

g

Oth

erS

ourc

esof

Bia

sP

atie

ntO

utco

mes

Pro

cess

esof

Car

eH

ealth

Ser

vice

Out

com

es

Ran

dom

Seq

uenc

eG

ener

atio

nA

lloca

tion

Con

ceal

men

t

Girg

iset

al,23

2009

No

over

alli

nter

vent

ion

effe

ctw

asob

serv

ed.P

hysi

calf

unct

ioni

ngw

assi

gnifi

cant

lyim

prov

edat

the

third

tele

phon

ein

terv

iew

for

parti

cipa

nts

inth

ete

leph

one

case

wor

kerg

roup

(P�

.01)

and

ther

ew

asa

trend

tow

ard

few

erpa

rtici

pant

sw

ithun

met

need

s(P

�.0

7).

Pat

ient

sin

the

tele

phon

eca

sew

orke

rgr

oup

wer

em

ore

likel

yto

have

indi

cate

dis

sues

ofne

eddi

scus

sed

(P�

.001

),re

ferr

als

mad

e(P

�.0

01),

and

stro

ngag

reem

ent

that

the

inte

rven

tion

impr

oved

com

mun

icat

ion

with

thei

rhe

alth

care

team

(P�

.001

).

—Y

esLo

wU

ncle

arH

igh

Hig

hLo

wU

ncle

arH

igh

Hoe

kstr

aet

al,25

2006

Att

he2-

mfo

llow

-up,

the

prev

alen

ceof

sym

ptom

sw

aslo

wer

inth

eEG

(pre

vale

ntdi

ffere

nces

2.1%

–24.

3%)f

orni

neof

10sy

mpt

oms

(exc

ept

coug

hing

).C

onst

ipat

ion

and

vom

iting

wer

esi

gnifi

cant

lyle

sspr

eval

enti

nEG

.Sev

erity

offa

tigue

,lac

kof

appe

tite,

shor

tnes

sof

brea

th,a

ndna

usea

was

low

erin

the

EG(n

otsi

gnifi

cant

).N

oEG

/CG

diffe

renc

esin

seve

rity

ofpa

in,

coug

hing

,sle

eple

ssne

ss,a

nddi

arrh

ea).

——

No

Hig

hU

ncle

arH

igh

Low

Low

Unc

lear

Hig

h

Kea

rney

etal

,41

2009

CG

had

sign

ifica

ntly

mor

ere

ports

offa

tigue

(P�

.04)

and

sign

ifica

ntly

few

erre

ports

ofha

nd-fo

otsy

ndro

me

(P�

.03)

than

EG.N

oEG

/CG

diffe

renc

esin

repo

rtsof

vom

iting

/nau

sea,

diar

rhea

,ors

ore

mou

th/th

roat

.N

oEG

/CG

diffe

renc

esin

seve

rity

and

dist

ress

ofsy

mpt

oms,

with

the

exce

ptio

nof

high

erse

verit

y(P

�.0

3)an

ddi

stre

ss(P

�.0

3)of

hand

-foot

synd

rom

ein

EG.

——

No

Low

Low

Hig

hLo

wH

igh

Low

Hig

h

Klin

kham

mer

-S

chal

keet

al,26

2012

At6

m,7

1%of

patie

nts

inC

Gsh

owed

dise

ased

QoL

inat

leas

ton

edi

men

sion

.In

the

EG,t

his

occu

rred

in56

%of

patie

nts

(P�

.048

).R

elat

ive

risk

was

redu

ced

21%

(95%

CI,

0to

37)

and

abso

lute

risk

was

redu

ced

15%

(95%

CI,

0.3

to29

).Th

eN

o.of

dise

ased

QoL

dim

ensi

ons

perp

atie

ntw

aslo

wer

inth

eEG

at6

m(P

�.0

35).

The

perc

ent

ofpa

tient

sw

ithze

roQ

oLin

atle

asto

nedi

men

sion

at6

mw

as15

%in

the

EGan

d25

%in

the

CG

(P�

.124

).

At

3m

,co

ping

stra

tegi

esw

ere

appl

ied

mor

eof

ten

but

not

sign

ifica

ntly

mor

ein

the

EG

than

the

CG

(P�

.055

).S

igni

fican

tlym

ore

psyc

hoth

erap

yw

asgi

ven

tow

omen

inth

eE

G(P

�.0

05)

but

the

oppo

site

was

true

for

phys

ioth

erap

yin

the

CG

.A

t6

m,

the

resu

ltsw

ere

muc

hm

ore

sim

ilar

inth

eE

Gan

dC

G.

—Y

esLo

wLo

wH

igh

Hig

hLo

wLo

wLo

w

(con

tinue

don

follo

win

gpa

ge)

Kotronoulas et al



Tabl

e2.

Mai

nFi

ndin

gsan

dA

sses

smen

tof

Ris

kof

Bia

sin

the

20R

CTs

and

Four

Non

-RC

TsId

entifi

edfo

rTh

isR

evie

w(c

ontin

ued)

Aut

hor

and

Yea

r

Mai

nS

tudy

Find

ings

�

Gui

delin

esW

ere

Use

dto

Gui

deC

linic

ian

Res

pons

e

Ris

kof

Bia

s†

Sel

ectio

nB

ias

Per

form

ance

Bia

s:B

lindi

ngP

artic

ipan

tsan

dP

erso

nnel

Det

ectio

nB

ias:

Blin

ding

ofO

utco

me

Ass

essm

ent

Att

ritio

nB

ias:

Inco

mpl

ete

Out

com

eD

ata

Rep

ortin

gB

ias:

Sel

ectiv

eO

utco

me

Rep

ortin

g

Oth

erS

ourc

esof

Bia

sP

atie

ntO

utco

mes

Pro

cess

esof

Car

eH

ealth

Ser

vice

Out

com

es

Ran

dom

Seq

uenc

eG

ener

atio

nA

lloca

tion

Con

ceal

men

t

Kor

nblit

het

al,42

2006

EG

had

sign

ifica

ntly

low

eran

xiet

yan

dde

pres

sion

at6

m(P

�.0

01).

No

diff

eren

ces

onps

ycho

logi

cald

istr

ess,

QoL

,or

com

orbi

ditie

sin

terf

erin

gw

ithfu

nctio

ning

.S

igni

fican

tlym

ore

patie

nts

inth

eE

Gha

dsc

ores

abov

ecu

tof

ffo

rde

pres

sion

/anx

iety

at6

m(4

2%v

24%

;P

�.0

41).

No

sign

ifica

ntE

G/C

Gdi

ffer

ence

sin

perc

ent

ofph

ysic

alsy

mpt

omal

erts

.N

odi

ffer

ence

sin

over

all

satis

fact

ion

with

inte

rven

tion

(goo

d/ex

celle

nt,

88%

v74

%).

Sig

nific

antly

few

erpa

tient

sin

the

EG

rate

dth

ein

terv

entio

nve

ry/e

xtre

mel

yhe

lpfu

lin

copi

ngw

ithan

impo

rtan

tpr

oble

m(P

�.0

18).

No

EG

/CG

diff

eren

ces

inus

eof

men

talh

ealth

serv

ices

at6

m(9

%v

12%

).

No

Unc

lear

Unc

lear

Hig

hH

igh

Hig

hU

ncle

arU

ncle

ar

Mau

nsel

let

al,27

1996

Par

ticip

ants

’ps

ycho

logi

cal

dist

ress

leve

lsde

crea

sed

over

the

stud

ype

riod

(P�

.001

),bu

tno

EG

/CG

diff

eren

ces.

No

EG

/CG

diff

eren

ces

inph

ysic

alhe

alth

,fu

nctio

nals

tatu

s,so

cial

and

leis

ure

activ

ities

,re

turn

tow

ork,

orm

arita

lsa

tisfa

ctio

n.

No

EG

/CG

diff

eren

ces

inth

em

ean

No.

ofso

cial

wor

ker

cont

acts

.C

Gan

dE

Gw

ere

very

sim

ilar

into

tal

inte

rven

tion

time,

prop

ortio

nof

cont

acts

cond

ucte

din

pers

on,

and

mea

ndu

ratio

nof

cont

acts

cond

ucte

din

pers

onan

dby

tele

phon

edu

ring

the

base

line

perio

d.U

seof

psyc

hoso

cial

serv

ices

,m

edic

alco

nsul

tatio

ns,

orot

her

patie

ntin

itiat

ives

that

mig

htim

prov

equ

ality

oflif

edi

dno

tdi

ffer

betw

een

grou

ps.

The

mea

nN

o.of

visi

tsw

as2.

4an

d6.

1am

ong

CG

and

EG

patie

nts,

resp

ectiv

ely,

repr

esen

ting

48.9

and

119.

6m

inof

soci

alw

orke

rco

ntac

t.

No

Low

Low

Hig

hU

ncle

arH

igh

Unc

lear

Hig

h

McL

achl

anet

al,28

2001

No

EG

/CG

diff

eren

ces

inch

ange

sin

psyc

holo

gica

lor

heal

thin

form

atio

nne

eds,

QoL

,or

psyc

hoso

cial

func

tioni

ngbe

twee

nth

eba

selin

ean

dfo

llow

-up

asse

ssm

ents

.Fo

rth

esu

bgro

upof

mod

erat

ely/

seve

rely

depr

esse

dpa

tient

s,th

ere

was

asi

gnifi

cant

redu

ctio

nin

depr

essi

onfo

rth

eE

Gre

lativ

eto

the

CG

atth

e6-

mas

sess

men

t(P

�.0

01).

No

EG

/CG

diff

eren

ces

(P�

.36)

inco

nsul

tatio

ntim

es(1

7.7

min

v16

.4m

in)

orle

vels

ofsa

tisfa

ctio

n(P

�.0

5).

For

CG

vE

Gpa

tient

s,th

epe

rcen

tof

patie

nts

indi

catin

gth

eir

leve

lof

satis

fact

ion

was

95%

v98

%fo

rnu

rsin

gca

re,

98%

v98

%fo

rm

edic

alca

re,

91%

v96

%fo

rin

form

atio

nre

ceiv

edab

out

thei

rill

ness

and

trea

tmen

t,an

d98

%v

99%

for

over

all

satis

fact

ion

with

the

care

rece

ived

.

—Y

esLo

wLo

wH

igh

Low

Low

Unc

lear

Hig

h

(con

tinue

don

follo

win

gpa

ge)




Tabl

e2.

Mai

nFi

ndin

gsan

dA

sses

smen

tof

Ris

kof

Bia

sin

the

20R

CTs

and

Four

Non

-RC

TsId

entifi

edfo

rTh

isR

evie

w(c

ontin

ued)

Aut

hor

and

Yea

r

Mai

nS

tudy

Find

ings

�

Gui

delin

esW

ere

Use

dto

Gui

deC

linic

ian

Res

pons

e

Ris

kof

Bia

s†

Sel

ectio

nB

ias

Per

form

ance

Bia

s:B

lindi

ngP

artic

ipan

tsan

dP

erso

nnel

Det

ectio

nB

ias:

Blin

ding

ofO

utco

me

Ass

essm

ent

Att

ritio

nB

ias:

Inco

mpl

ete

Out

com

eD

ata

Rep

ortin

gB

ias:

Sel

ectiv

eO

utco

me

Rep

ortin

g

Oth

erS

ourc

esof

Bia

sP

atie

ntO

utco

mes

Pro

cess

esof

Car

eH

ealth

Ser

vice

Out

com

es

Ran

dom

Seq

uenc

eG

ener

atio

nA

lloca

tion

Con

ceal

men

t

Mill

set

al,29

2009

Onl

ya

smal

lbut

cons

iste

ntdi

ffer

ence

inQ

oLw

asfo

und

betw

een

EG

and

CG

.Th

eE

Gha

da

poor

erQ

oLin

man

ydo

mai

ns.

Two

diff

eren

tQ

oLsu

mm

ary

scor

esin

dica

ted

ast

atis

tical

lysi

gnifi

cant

betw

een-

grou

pdi

ffer

ence

.

Onl

y23

%of

the

diar

ygr

oup

stat

edth

atth

eyha

dsh

ared

thei

rdi

ary

with

any

heal

thpr

ofes

sion

al.

No

inte

rven

tion

effe

cts

inco

mm

unic

atio

n,sa

tisfa

ctio

nw

ithca

re,

orth

edi

scus

sion

ofpa

tient

prob

lem

s.E

Gdi

scus

sed

few

erto

pics

with

heal

thpr

ofes

sion

als

than

the

CG

(not

sign

ifica

nt).

Bot

hgr

oups

repo

rted

high

leve

lsof

satis

fact

ion

with

thei

rca

re,

with

nosi

gnifi

cant

asso

ciat

ions

iden

tified

.

—N

oLo

wLo

wH

igh

Low

Low

Unc

lear

Low

Nic

klas

son

etal

,3020

13—

Issu

esre

gard

ing

emot

iona

lfu

nctio

ning

wer

em

ore

freq

uent

lydi

scus

sed

inth

eE

Gby

doct

ors

orpa

tient

sta

ken

toge

ther

(P�

.015

).N

oE

G/C

Gdi

ffer

ence

sin

phys

ical

/rol

e,so

cial

,co

gniti

vefu

nctio

ning

,or

glob

alhe

alth

.P

ain,

dysp

nea,

fatig

ue,

and

anor

exia

wer

eso

mew

hat

mor

efr

eque

ntly

disc

usse

din

the

EG

(not

sign

ifica

nt).

Med

ical

/tec

hnic

alst

atem

ents

wer

em

ore

freq

uent

lyra

ised

inth

eC

G(P

�.0

5).

Leng

thof

doct

or-

patie

ntco

nver

satio

nsw

assi

mila

rin

the

EG

and

CG

(P�

.77)

.N

o.of

diag

nost

ican

dth

erap

eutic

inte

rven

tions

for

emot

iona

land

soci

alco

ncer

nsw

ashi

gher

inth

eE

G.

Pla

nned

outp

atie

ntvi

sits

wer

esi

mila

rbe

twee

nE

Gan

dC

G(3

27v

323)

.

No

Unc

lear

Hig

hH

igh

Low

Low

Unc

lear

Low

Ros

enbl

oom

etal

,3120

07N

ost

atis

tical

lysi

gnifi

cant

diff

eren

ces

acro

ssth

eth

ree

stud

yco

nditi

ons

inQ

oLov

ertim

e(P

�.0

5).

For

all

patie

nts,

QoL

esse

ntia

llydi

dno

tch

ange

over

the

cour

seof

the

stud

y.

No

sign

ifica

ntdi

ffer

ence

sac

ross

the

thre

est

udy

cond

ition

sin

gene

ral

satis

fact

ion

and

satis

fact

ion

with

com

mun

icat

ion

over

time

(P�

.05)

.Fo

ral

lpa

tient

s,sa

tisfa

ctio

nes

sent

ially

did

not

chan

geov

erth

eco

urse

ofth

est

udy.

No

sign

ifica

ntgr

oup

diff

eren

ces

(P�

.05)

incl

inic

altr

eatm

ent

chan

ges

betw

een

the

thre

eco

nditi

ons.

—N

oU

ncle

arU

ncle

arH

igh

Hig

hLo

wU

ncle

arH

igh

(con

tinue

don

follo

win

gpa

ge)

Kotronoulas et al



Tabl

e2.

Mai

nFi

ndin

gsan

dA

sses

smen

tof

Ris

kof

Bia

sin

the

20R

CTs

and

Four

Non

-RC

TsId

entifi

edfo

rTh

isR

evie

w(c

ontin

ued)

Aut

hor

and

Yea

r

Mai

nS

tudy

Find

ings

�

Gui

delin

esW

ere

Use

dto

Gui

deC

linic

ian

Res

pons

e

Ris

kof

Bia

s†

Sel

ectio

nB

ias

Per

form

ance

Bia

s:B

lindi

ngP

artic

ipan

tsan

dP

erso

nnel

Det

ectio

nB

ias:

Blin

ding

ofO

utco

me

Ass

essm

ent

Att

ritio

nB

ias:

Inco

mpl

ete

Out

com

eD

ata

Rep

ortin

gB

ias:

Sel

ectiv

eO

utco

me

Rep

ortin

g

Oth

erS

ourc

esof

Bia

sP

atie

ntO

utco

mes

Pro

cess

esof

Car

eH

ealth

Ser

vice

Out

com

es

Ran

dom

Seq

uenc

eG

ener

atio

nA

lloca

tion

Con

ceal

men

t

Rul

and

etal

,32

2010

Sym

ptom

dist

ress

inth

eE

Gde

crea

sed

sign

ifica

ntly

over

time

in11

(58%

)of

19sy

mpt

om/p

robl

emca

tego

ries

vtw

o(1

0%)

for

the

CG

.

Sig

nific

antly

mor

esy

mpt

oms

wer

ead

dres

sed

inth

eE

Gpa

tient

char

tsv

thos

eof

the

CG

.N

eed

for

sym

ptom

man

agem

ent

supp

ort

over

time

also

decr

ease

dsi

gnifi

cant

lym

ore

for

the

EG

than

the

CG

in13

(68%

)sy

mpt

omca

tego

ries.

—N

oLo

wH

igh

Hig

hH

igh

Low

Unc

lear

Hig

h

Sar

na,33

1998

Sym

ptom

dist

ress

scor

esof

the

CG

wer

ehi

gher

than

scor

esof

the

EG

(P�

.001

).C

hem

othe

rapy

stat

usan

dgr

oup

assi

gnm

ent

wer

ebo

thst

rong

pred

icto

rsof

dist

ress

scor

es.

The

no-c

hem

othe

rapy

subg

roup

show

edgr

eate

rle

vels

ofdi

stre

ssth

anth

ech

emot

hera

pysu

bgro

upw

ithth

eC

Gan

dE

Ggr

oups

.

——

No

Unc

lear

Unc

lear

Hig

hU

ncle

arH

igh

Hig

hH

igh

Take

uchi

etal

,3520

11;

Vel

ikov

aet

al,38

2004

;an

dV

elik

ova

etal

,39

2010

‡

Pat

ient

sin

the

EG

and

atte

ntio

n-co

ntro

lgro

upha

dbe

tter

QoL

than

the

CG

(P�

.006

and

P�

.01,

resp

ectiv

ely)

,bu

tth

eE

Gan

dat

tent

ion-

cont

rol

grou

psw

ere

not

sign

ifica

ntly

diff

eren

t(P

�.8

0).

Ala

rger

prop

ortio

nof

inte

rven

tion

patie

nts

show

edcl

inic

ally

mea

ning

fuli

mpr

ovem

ent

inQ

oL.

Mor

efr

eque

ntdi

scus

sion

ofch

roni

cno

nspe

cific

sym

ptom

s(P

�.0

3)in

the

EG

,w

ithou

tpr

olon

ging

enco

unte

rs.

No

effe

cton

patie

ntm

anag

emen

t(P

�.6

0).

Dis

cuss

ion

topi

csw

ere

pred

omin

antly

rais

edby

patie

nts/

rela

tives

,re

gard

less

ofgr

oup.

Clin

icdi

scus

sion

sw

ere

asso

ciat

edw

ithse

verit

yof

repo

rted

sym

ptom

s,bu

tno

tw

ithpa

tient

-rep

orte

dfu

nctio

nal

conc

erns

.E

Gpa

tient

sra

ted

thei

rco

ntin

uity

ofca

reas

bett

erth

anth

eC

Gin

term

sof

com

mun

icat

ion

(P�

.03)

.P

atie

nts’

eval

uatio

nsof

the

inte

rven

tion

wer

epo

sitiv

e.

—N

oU

ncle

arU

ncle

arH

igh

Unc

lear

Hig

hU

ncle

arH

igh

Trow

brid

geet

al,37

1997

No

sign

ifica

ntE

G/C

Gdi

ffer

ence

sin

asse

ssm

ents

ofpa

in,

pain

regi

men

s,an

dre

lief

rece

ived

atth

e4-

wee

kfo

llow

-up.

Sig

nific

ant

EG

/CG

diff

eren

ces

inph

ysic

ians

’pa

tter

nsof

pres

crib

ing

anal

gesi

cs(2

5%v

14%

;P

�.0

16).

No

sign

ifica

ntdi

ffer

ence

sin

the

perc

enta

geof

patie

nts

unde

rtre

ated

for

pain

(38%

v35

%;

P�

.05)

.

—N

oH

igh

Hig

hH

igh

Hig

hU

ncle

arH

igh

Hig

h

(con

tinue

don

follo

win

gpa

ge)




Tabl

e2.

Mai

nFi

ndin

gsan

dA

sses

smen

tof

Ris

kof

Bia

sin

the

20R

CTs

and

Four

Non

-RC

TsId

entifi

edfo

rTh

isR

evie

w(c

ontin

ued)

Aut

hor

and

Yea

r

Mai

nS

tudy

Find

ings

�

Gui

delin

esW

ere

Use

dto

Gui

deC

linic

ian

Res

pons

e

Ris

kof

Bia

s†

Sel

ectio

nB

ias

Per

form

ance

Bia

s:B

lindi

ngP

artic

ipan

tsan

dP

erso

nnel

Det

ectio

nB

ias:

Blin

ding

ofO

utco

me

Ass

essm

ent

Att

ritio

nB

ias:

Inco

mpl

ete

Out

com

eD

ata

Rep

ortin

gB

ias:

Sel

ectiv

eO

utco

me

Rep

ortin

g

Oth

erS

ourc

esof

Bia

sP

atie

ntO

utco

mes

Pro

cess

esof

Car

eH

ealth

Ser

vice

Out

com

es

Ran

dom

Seq

uenc

eG

ener

atio

nA

lloca

tion

Con

ceal

men

t

Non

-RC

TsB

oyes

etal

,18

2006

Pat

ient

sin

the

EG

with

ade

bilit

atin

gsy

mpt

omat

visi

t2

wer

ele

sslik

ely

tore

port

ade

bilit

atin

gsy

mpt

omat

visi

t3

com

pare

dw

ithC

G(P

�.0

4).

No

EG

/CG

diff

eren

ces

inch

ange

inan

xiet

y(P

�.0

9)an

dde

pres

sion

scor

es(P

�.2

0).

No

sign

ifica

ntE

G/C

Gdi

ffer

ence

sin

chan

gein

aver

age

No.

ofm

oder

ate

orhi

ghps

ycho

logi

caln

eeds

repo

rted

over

time

(P�

.82)

.

For

patie

nts,

the

inte

rven

tion

was

easy

toco

mpl

ete,

and

they

wou

ldbe

will

ing

toco

mpl

ete

the

surv

eyea

chtim

eth

eyvi

site

dth

eon

colo

gist

.O

nly

thre

eE

Gpa

tient

sre

port

edth

atth

eir

onco

logi

stdi

scus

sed

the

feed

back

repo

rtw

ithth

em.

Hal

fof

the

med

ical

onco

logi

sts

(n�

2)re

port

edth

atth

eydi

scus

sed

the

feed

back

dire

ctly

with

thei

rpa

tient

s.

—N

oH

igh

Hig

hH

igh

Low

Hig

hU

ncle

arLo

w

Hila

rius

etal

,24

2008

No

sign

ifica

ntef

fect

sw

ere

foun

din

chan

ges

inQ

oLov

ertim

e.

QoL

-rel

ated

topi

csdi

scus

sed

mor

efr

eque

ntly

inth

eE

G(P

�.0

2).

Nur

ses’

awar

enes

sof

patie

nts’

leve

lsof

daily

activ

ity,

pain

,an

dov

eral

lQ

oLw

assi

gnifi

cant

lybe

tter

inth

eE

G.

The

mea

nN

o.of

QoL

-rel

ated

nota

tions

inth

em

edic

alre

cord

sw

ashi

gher

inth

eE

G(P

�.0

5).

Mod

est

effe

cts

wer

eob

serv

edin

patie

ntm

anag

emen

t;no

sign

ifica

ntef

fect

sin

patie

ntsa

tisfa

ctio

nov

ertim

e.

—N

oH

igh

Hig

hH

igh

Hig

hH

igh

Unc

lear

Unc

lear

Taen

zer

etal

,34

2000

—In

the

EG

,m

ore

QoL

issu

esid

entifi

edby

the

patie

ntw

ere

addr

esse

ddu

ring

the

clin

icap

poin

tmen

tth

anin

the

CG

(P�

.01)

.M

argi

nally

mor

eca

tego

ries

wer

ech

arte

dan

da

tren

dto

war

dm

ore

actio

nsbe

ing

take

nw

asre

cord

edin

the

EG

.P

atie

nts

repo

rted

bein

geq

ually

and

high

lysa

tisfie

dre

gard

less

ofst

udy

grou

p.

—N

oH

igh

Hig

hH

igh

Unc

lear

Low

Unc

lear

Hig

h

(con

tinue

don

follo

win

gpa

ge)

Kotronoulas et al



Tabl

e2.

Mai

nFi

ndin

gsan

dA

sses

smen

tof

Ris

kof

Bia

sin

the

20R

CTs

and

Four

Non

-RC

TsId

entifi

edfo

rTh

isR

evie

w(c

ontin

ued)

Aut

hor

and

Yea

r

Mai

nS

tudy

Find

ings

�

Gui

delin

esW

ere

Use

dto

Gui

deC

linic

ian

Res

pons

e

Ris

kof

Bia

s†

Sel

ectio

nB

ias

Per

form

ance

Bia

s:B

lindi

ngP

artic

ipan

tsan

dP

erso

nnel

Det

ectio

nB

ias:

Blin

ding

ofO

utco

me

Ass

essm

ent

Att

ritio

nB

ias:

Inco

mpl

ete

Out

com

eD

ata

Rep

ortin

gB

ias:

Sel

ectiv

eO

utco

me

Rep

ortin

g

Oth

erS

ourc

esof

Bia

sP

atie

ntO

utco

mes

Pro

cess

esof

Car

eH

ealth

Ser

vice

Out

com

es

Ran

dom

Seq

uenc

eG

ener

atio

nA

lloca

tion

Con

ceal

men

t

Thew

eset

al,36

2009

Par

ticip

ants

inth

esc

reen

edco

hort

repo

rted

sign

ifica

ntly

high

erle

vels

ofov

eral

lunm

etne

eds

(P�

.001

),ps

ycho

logi

caln

eeds

(P�

.02)

,in

form

atio

nne

eds

(P�

.02)

,an

dph

ysic

alan

dda

ilyliv

ing

need

s(P

�.0

4)co

mpa

red

with

the

unsc

reen

edco

hort

.N

odi

ffer

ence

son

sexu

ality

need

s.

Scr

eeni

ngdi

dno

tsi

gnifi

cant

lyin

crea

seth

era

teof

refe

rral

sto

psyc

hoso

cial

staf

fof

dist

ress

edin

divi

dual

s,bu

tre

duce

dtim

eto

refe

rral

.

—N

oH

igh

Unc

lear

Hig

hU

ncle

arU

ncle

arU

ncle

arU

ncle

ar

Abb

revi

atio

ns:

CG

,co

ntro

lgro

up;

EG

,ex

perim

enta

lgro

up;

m,

mon

ths;

min

,m

inut

es;

QoL

,qu

ality

oflif

e;R

CT,

rand

omiz

edco

ntro

lled

tria

l.�In

form

atio

non

effe

ctsi

zes

(whe

reca

lcul

atio

nw

aspe

rmitt

edby

avai

labi

lity

ofda

ta)

isav

aila

ble

inA

ppen

dix

Tabl

esA

2,A

3,an

dA

4(o

nlin

eon

ly).

†Spe

cific

expl

anat

ions

for

allr

atin

gsar

eav

aila

ble

from

the

auth

ors.

‡Art

icle

sar

eba

sed

onda

tafr

omth

esa

me

stud

y;di

ffer

ent

sam

ple

size

san

dou

tcom

esar

eev

alua

ted

inea

char

ticle

.




of intervention effects on actions taken as a result of PROM feedback becom-ing available to clinicians remains generally ambiguous (Appendix Table A4).No significant intervention effects were reported in the number of patientsreferred to psychosocial care19,20,36 or in clinical actions taken.21,24,31 Al-though at 3 months after the intervention women with breast cancer in theexperimental group were offered counseling and psychotherapy services moreoften, at 6 months this difference disappeared.26 When PROMs were used toincrease physician awareness of patients’ levels of pain, a significant change(d � 0.41) in analgesic prescription patterns was found to favor the experi-mental group.37 During treatment for chest malignancies, significantly morepatients in the experimental group received diagnostic and therapeutic servicesfor emotional and social concerns,30 but numbers of QoL-related actionstaken per patient were similar across study groups.34

Patient satisfaction with care and/or communication with team. Regard-less of study condition, patient remarks on satisfaction with care and/or com-munication with HPs were generally positive.19,21,24,28,29,31,34,39,40 Thougheight CTs19,24,28,29,31,34,40 failed to show significant intervention effects (Ap-pendix Table A4). In the studies in which postintervention gains were re-ported, the positive effects referred to greater satisfaction with emotionalsupport in the palliative chemotherapy context,21 greater satisfaction withpatients receiving follow-up from oncology nurses rather than general practi-tioners (though differences from usual care were not examined),23 and en-hanced communication with physicians in the outpatient setting comparedwith standard care.39

Patient outcomes discussed during consultation. Regardless of patients’cancer type, significant postintervention increases over time in the frequencyof discussions pertinent to patient outcomes during consultations were re-

corded.35,38 The odds of such outcomes being discussed seemed to depend onwhether these were reported at a level indicating a problem.17 Though emo-tional problems tend to be discussed more often during consultations in theexperimental group,19 social and sexual functioning issues may be those onwhich the intervention proves most effective.17 Still, the overall patient-physician communication may not significantly improve.19 In the lung cancerpopulation, significantly more symptoms were discussed and addressed dur-ing consultations,34 but intervention effects on QoL discussions fell short ofsignificance. Much greater intervention effects were reported in the context ofpalliative chemotherapy (Appendix Table A4),21 regarding overall communica-tion about dyspnea (d � 0.40 to 0.77)21,24; social functioning (d � 0.49) andfatigue (d � 0.38)21; and sleep problems (d � 0.66), constipation (d � 0.40),diarrhea (d � 0.67), and cognitive functioning (d � 0.66).24

HP acceptability/evaluation of intervention. Where addressed, interven-tion acceptability was moderate to high across all CTs (Table 2), with rates ofperceived usefulness ranging from less than 50% to 68%. HPs felt obtaining anoverall assessment of the patient was more helpful21,38,39 to identify issues ofconcern17,19,21,38 and to guide discussions with patients17-19,24 rather than incommunicating with patients17,19 and in managing and enhancing the careprovided.18,38 Yet, in two similar CTs, all physicians21 and nurses24 agreed thatthe intervention facilitated patient-clinician communication. The ability ofHPs to identify psychosocial concerns19,21 and address difficult subjects suchas sexuality issues24 was also enhanced. Although actual changes in HP com-munication styles may not be seen even following the intervention,19 physi-cians21,39 and nurses24 seem willing to continue using the PROM summary ineveryday practice. Nurses significantly more frequently found PROM

Table 3. Classification of Study Outcomes According to the Three Prespecified Outcome Categories (n � 24)

Patient Outcomes Processes of Care Health Service Outcomes

ClassificationNo. ofStudies % Classification

No. ofStudies % Classification

No. ofStudies %

Physical symptoms:prevalence and/orseverity18,22,25,32,37,40,41

7 29.2 Patient actual use of the interventionPROM25,29

2 8.3 Health services use/self-referrals20,22,42

3 12.5

QoL21-24,26,28,29,31,38,42 10 41.7 Duration of contacts withHPs17-19,27,28,30,38

7 29.2 Contact with HPs27,30 2 8.3

Psychologicalsymptoms18,20,23,27,28,42

6 25 Patient engagement in self-careactions27

1 4.2

Supportive careneeds18,22,23,32,36

5 20.8 Patient outcomes discussed duringconsultation17,19,21,24,29,30,34,35,38

8 33.3

Overall distress20,31,33 3 12.5 HP acceptability/evaluation ofintervention17-19,21,24,38,39

6 25.0

Overall physical health27,42 2 8.3 Patient satisfaction withcare/communication with treatingteam19,21,23,24,28,29,31,34,37,39,40

11 45.8

Working hours27 1 4.2 Patient outcomes addressed inpatient records32,34

2 8.3

Social support27 1 4.2 Medical decisions made/advicegiven/changes intreatment/referrals made19-

21,23,24,26,30,31,34,36,37

11 45.8

Social activity27 1 4.2 HP use of PROM information38,39 1 4.2Physical activity27 1 4.2 HP satisfaction with encounter with

the patient211 4.2

Marital satisfaction27 1 4.2 HP awareness of patientoutcomes21,24

2 8.3

Patient satisfaction withintervention19,21,24,36,39,40,42

7 29.2

Impact of referrals on patientoutcomes20

1 4.2

Perceived continuity andcoordination of care39

1 4.2

Timing of referrals19,36 2 8.3

Abbreviations: HP, health professional; PROM, patient-reported outcome measure; QoL, quality of life.

Kotronoulas et al



interventions beneficial17 and felt that use of relevant information resulted inmore efficient use of their time.24

Patient satisfaction with intervention. Overall satisfaction with interven-tion was evident for at least 80% of patients.40,42 The PROM interventionswere seen as easy to use40 and a useful way for patients to describe theirsituation39 and communicate important information to HPs.19 Patients ex-pressed their willingness to continue using it in routine care.39,40 However, inthe Kornblith et al42 CT, percentages of patients rating the PROM interventionas very or extremely helpful in coping with an important problem were notablylow and favored the control rather than the experimental group (37% v 14%;d � 0.69). More than 83% of patients regarded the PROM content importantfor them and its use necessary for all patients receiving treatment.19,36 More-over, almost all patients (93%) appreciated having been asked about theiremotional well-being during treatment.36 In the palliative care setting, patientsagreed that the summary profile enhanced their physician’s or nurse’s aware-ness of their health problems (79% to 89%), and that it would be useful as astandard part of their consultations (87% to 99%).21,24

HP awareness of patient outcomes. In the context of palliative chemo-therapy, no intervention effects were reported on the magnitude of patient-physician agreement about patients’ physical, emotional, and social well-beingand daily activities (d � 0.09 to 0.50; Appendix Table A4).21 The only excep-tion was greater agreement in ratings of social functioning in the experimentalgroup, but this applied only to the subgroup of patients who reportedmoderate-to-severe problems.21 Oncology nurses’ awareness of daily activi-ties, pain, and QoL was significantly higher in the experimental group duringthe fourth patient visit.24 Positive intervention effects were reported in patientcare documentation in the medical records of patients being treated for hema-tologic malignancies32 and in the number of QoL issues charted in records ofpatients with lung cancer.34

Timing of referrals. One RCT revealed that PROM feedback resulted insignificantly earlier postconsultation referral of patients in the experimentalversus the control group by an average of three weeks.19 In a sequential cohorttrial of patient-distress screening, average time to referral in the unscreenedcohort was 14 days compared with a considerably earlier referral of only 5 daysin the screened cohort.36

Health Services Outcomes

Only five CTs explored the effects of the routine use of PROMs on HSOs(Table 3; Appendix Table A5), namely, numbers of patients making use ofhealth services20,22,42 and frequency of contacts with health professionals.27,30

Ganz et al22 reported only minimal use of services after referral to psychosocialcare in women with breast cancer; whereas prevalence of cases in whichpatients sought professional help was similar irrespective of study groupamong newly diagnosed patients with lung cancer and breast cancer.20 Amongpatients with advanced breast, colorectal, or prostate cancer, use of mentalhealth services at 6 months after intervention was equally minimal regardlessof study condition (P � .34).42 In terms of frequency of patient-HP contacts,positive intervention effects were found among women with breast cancer27

but not among patients with chest malignancies.30

DISCUSSION

We found only tentative evidence regarding the effectiveness ofPROM interventions to improve the quality of care provided to pa-tients receiving active anticancer treatments. We used strict systematicmethods during identification12 and risk-of-bias appraisal13 of alltrials included here. We included 24 CTs, which investigated a widerange of outcomes, thus producing a disparate set of data and indicat-ing lack of consensus around the role of PROMs and the range ofoutcome measures in clinical practice. Evidence suggests that, irre-spective of the context of chronic illness, the impact of PROMs on POsis weak.9,44 Where possible, we calculated ES in an attempt to quantifythe magnitude of these effects, and our findings indicate inconsisten-cies in the overall significance (statistical or clinical) and low-to-

moderate intervention effectiveness. Importantly, efficacy of the CTsreviewed seems low, confirming findings from previous reviews.5,9,44

Contrary to the limited evaluation of HSOs, PoCs were the mostfrequently investigated outcomes in our sample of trials. Mixed find-ings emerged regarding medical decisions made or actions taken byHPs as a result of the availability of PROM data. Changes in HPpractices fell short of significance and, where such changes were doc-umented,30,37 the associated ES were still small. It is unclear whetherlimited referral options, additional subjective HP assessments, orother health care�related factors influenced the use of PROMs inpractice. Patient satisfaction with care did not improve significantly,possibly owing to the presence of ceiling effects. Moreover, achievableimprovements in patient communication with HPs, especially regard-ing emotional health issues, were documented, but ES were quitesmall. Somewhat greater ES can be proposed with regard to the actualdiscussion of POs during consultations, particularly physical symp-toms, but not necessarily around supportive care needs.19

Fewer than 30% of the CTs addressed the important question ofwhether the use of PROM interventions appeals to patients and HPs.Though HPs may view PROMs as useful toward a more comprehen-sive or systematic assessment, communication is not always enhanced.In addition, there is still limited (albeit positive) evidence aboutwhether HPs wish PROMs to become routine practice. Whether pa-tients can comply with the systematic use of PROMs during treatmentand encounters with the clinical team is equally unclear. Despitelimited evidence, including electronic systems to enhance data collec-tion and management, as well as use of clinical algorithms to supportclinicians in the management of identified areas for intervention,might potentially increase adherence to and acceptability of PROM-enhanced clinical assessments.

Current data also suggest that patient physical symptoms anddistress may be more amenable to improvement after PROM inter-ventions than QoL, supportive care needs, or psychological symp-toms. Even with the exception of the few studies that examined the useof health services by patients or contacts with HPs, important aspectsof an intervention’s applicability, such as patient safety or cost-effectiveness and cost-efficiency, are yet to be included as potential endpoints to encourage policy makers to consider making changes in theway cancer care is provided. Despite this lack of evidence, the Depart-ment of Health in England is aiming to extend the use of PROMs in awider range of conditions in that country’s National Health Service,45

which would include cancer care.Finally, measurement bias interfering with the effects of PROM

interventions documented in this review should also be considered.Arguably, not all tools used in the delivery of interventions wereoriginally developed as PROMs, which might have affected the reli-ability of reported outcomes and their subsequent interpretation. Inaddition, the psychometric robustness of the PROMs used to deliverand/or evaluate intervention effects is questionable and might haveinterfered with its ability to capture the actual magnitude of sucheffects. Similar comments can be made regarding sources of bias, suchas absence of randomization or uncertainty about whether cliniciansdid use information generated by PROMs during consultations,which may have further affected the trials’ internal and external valid-ity and adversely affected credibility of available evidence.

Our search strategy was purposefully inclusive, with an aim toinclude all relevant literature. However, it was limited to the mostcommon bibliographic databases, as well as to peer-reviewed articles




and reports published in the English language only. In addition, thegray literature was not searched. Owing to the vast heterogeneity in thestudies included, a meta-analytic synthesis was not feasible. Unavail-ability of data also prevented us from calculating ES for some of theincluded studies. However, such cases were equally distributed acrossstatistically significant and nonsignificant findings or across the differ-ent outcome categories; hence, we are confident that the associatedreporting bias has not greatly affected our conclusions.

More research is necessary on the effects of PROM interventionson health outcomes across different types of cancers and treatmentmodalities. The use of PROMs in clinical practice seems to be mosteffective in increasing patient satisfaction with communication aboutemotional concerns. Discussion of POs during consultations mayincrease and, in some studies, is associated with improved symptomcontrol, increased supportive care measures, and patient satisfaction.Additional patient-related outcomes could be usefully addressed infuture trials, including perceived self-care self-efficacy, social activity,work limitations, or survival. Patients and HPs are willing to engage inthe routine use of PROMs during anticancer treatment. However, it isparamount that PROM intervention implementation is effective andincorporates strategies that increase patient adherence to the actualuse of PROMs and HP engagement in the active incorporation ofPROM feedback during encounters with patients.44 Consensus is alsorequired on the standardization of PROMs to be used in future trials.Finally, dedicated research is required to support the cost-effective useof PROMs in clinical practice regarding patient safety, clinician bur-den, and health-services usage. This is an important area of consider-ation, particularly in times of increasing demands on health care.

AUTHORS’ DISCLOSURES OF POTENTIAL CONFLICTSOF INTEREST

Although all authors completed the disclosure declaration, the followingauthor(s) and/or an author’s immediate family member(s) indicated afinancial or other interest that is relevant to the subject matter underconsideration in this article. Certain relationships marked with a “U” arethose for which no compensation was received; those relationships markedwith a “C” were compensated. For a detailed description of the disclosurecategories, or for more information about ASCO’s conflict of interest policy,please refer to the Author Disclosure Declaration and the Disclosures ofPotential Conflicts of Interest section in Information for Contributors.Employment or Leadership Position: None Consultant or AdvisoryRole: None Stock Ownership: None Honoraria: None ResearchFunding: Nora Kearney, Philips HealthCare Expert Testimony:None Patents, Royalties, and Licenses: None Other Remuneration:None

AUTHOR CONTRIBUTIONS

Conception and design: Grigorios Kotronoulas, Nora Kearney,Roma MaguireCollection and assembly of data: Grigorios Kotronoulas, RomaMaguire, Alison Harrow, David Di Domenico, Suzanne Croy,Stephen MacGillivrayData analysis and interpretation: All authorsManuscript writing: All authorsFinal approval of manuscript: All authors

REFERENCES

1. Siegel R, DeSantis C, Virgo K, et al: Cancertreatment and survivorship statistics, 2012. CA Can-cer J Clin 62:220-241, 2012

2. Cheng KK, Yeung RM: Impact of mood distur-bance, sleep disturbance, fatigue and pain amongpatients receiving cancer therapy. Eur J Cancer Care(Engl) 22:70-78, 2013

3. Cleeland CS: Symptom burden: Multiplesymptoms and their impact as patient-reported out-comes. J Natl Cancer Inst Monogr 16-21, 2007

4. Trotti A, Colevas AD, Setser A, et al: Patient-reported outcomes and the evolution of adverseevent reporting in oncology. J Clin Oncol 25:5121-5127, 2007

5. Valderas JM, Alonso J: Patient reported out-come measures: A model-based classification sys-tem for research and clinical practice. Qual Life Res17:1125-1135, 2008

6. Espallargues M, Valderas JM, Alonso J: Pro-vision of feedback on perceived health status tohealth care professionals: A systematic review of itsimpact. Med Care 38:175-186, 2000

7. Gilbody SM, House AO, Sheldon T: Routineadministration of Health Related Quality of Life(HRQoL) and needs assessment instruments toimprove psychological outcome: A systematic re-view. Psychol Med 32:1345-1356, 2002

8. Greenhalgh J, Meadows K: The effectivenessof the use of patient-based measures of health inroutine practice in improving the process and out-comes of patient care: A literature review. J EvalClin Pract 5:401-416, 1999

9. Luckett T, Butow PN, King MT: Improvingpatient outcomes through the routine use of patient-reported data in cancer clinics: Future directions.Psychooncology 18:1129-1138, 2009

10. Marshall S, Haywood K, Fitzpatrick R: Impactof patient-reported outcome measures on routinepractice: A structured review. J Eval Clin Pract12:559-568, 2006

11. Mitchell AJ, Waller A, Carlson LE: Implement-ing a screening programme for distress in cancersettings: Science and practice. Psicooncologia9:259-275, 2012

12. Moher D, Liberati A, Tetzlaff J, et al: Preferredreporting items for systematic reviews and meta-analyses: The PRISMA statement. Ann Intern Med151:264-269, W64, 2009

13. Higgins JP, Altman DG, Gotzsche PC, et al:The Cochrane Collaboration’s tool for assessing riskof bias in randomised trials. BMJ 343:d5928, 2011

14. Lipsey MW, Wilson D: Practical Meta-Analysis (ed 1). Thousand Oaks, CA, Sage Publica-tions, 2000

15. Wilson BD: Practical Meta-Analysis Effect SizeCalculator. http://www.campbellcollaboration.org/escalc/html/EffectSizeCalculator-SMD-main.php

16. Cohen J: Statistical power analysis for thebehavioral sciences (ed 2). Hillsdale, NJ, LawrenceEarlbaum Associates, 1988

17. Berry DL, Blumenstein BA, Halpenny B, et al:Enhancing patient-provider communication with theelectronic self-report assessment for cancer: A ran-domized trial. J Clin Oncol 29:1029-1035, 2011

18. Boyes A, Newell S, Girgis A, et al: Doesroutine assessment and real-time feedback improvecancer patients’ psychosocial well-being? Eur J Can-cer Care (Engl) 15:163-171, 2006

19. Braeken AP, Kempen GI, Eekers D, et al: Theusefulness and feasibility of a screening instrumentto identify psychosocial problems in patients receiv-ing curative radiotherapy: A process evaluation.BMC Cancer 11:479, 2011

20. Carlson LE, Groff SL, Maciejewski O, et al:Screening for distress in lung and breast canceroutpatients: A randomized controlled trial. J ClinOncol 28:4884-4891, 2010

21. Detmar SB, Muller MJ, Schornagel JH, et al:Health-related quality-of-life assessments and patient-physician communication: A randomized controlled trial.JAMA 288:3027-3034, 2002

22. Ganz PA, Greendale GA, Petersen L, et al:Managing menopausal symptoms in breast cancersurvivors: Results of a randomized controlled trial.J Natl Cancer Inst 92:1054-1064, 2000

23. Girgis A, Breen S, Stacey F, et al: Impact oftwo supportive care interventions on anxiety, de-pression, quality of life, and unmet needs in patientswith nonlocalized breast and colorectal cancers.J Clin Oncol 27:6180-6190, 2009

24. Hilarius DL, Kloeg PH, Gundy CM, et al: Useof health-related quality-of-life assessments in dailyclinical oncology nursing practice: A communityhospital-based intervention study. Cancer 113:628-637, 2008

25. Hoekstra J, de Vos R, van Duijn NP, et al:Using the symptom monitor in a randomized con-trolled trial: The effect on symptom prevalence andseverity. J Pain Symptom Manage 31:22-30, 2006

26. Klinkhammer-Schalke M, Koller M, SteingerB, et al: Direct improvement of quality of life using atailored quality of life diagnosis and therapy path-way: Randomised trial in 200 women with breastcancer. Br J Cancer 106:826-838, 2012

Kotronoulas et al



http://www.campbellcollaboration.org/escalc/html/EffectSizeCalculator-SMD-main.php

http://www.campbellcollaboration.org/escalc/html/EffectSizeCalculator-SMD-main.php

27. Maunsell E, Brisson J, Deschenes L, et al:Randomized trial of a psychologic distress screeningprogram after breast cancer: Effects on quality oflife. J Clin Oncol 14:2747-2755, 1996

28. McLachlan SA, Allenby A, Matthews J, et al:Randomized trial of coordinated psychosocial inter-ventions based on patient self-assessments versusstandard care to improve the psychosocial function-ing of patients with cancer. J Clin Oncol 19:4117-4125, 2001

29. Mills ME, Murray LJ, Johnston BT, et al: Doesa patient-held quality-of-life diary benefit patientswith inoperable lung cancer? J Clin Oncol 27:70-77,2009

30. Nicklasson M, Elfstrom ML, Olofson J, et al:The impact of individual quality of life assessmenton psychosocial attention in patients with chestmalignancies: A randomized study. Support CareCancer 21:87-95, 2013

31. Rosenbloom SK, Victorson DE, Hahn EA, et al:Assessment is not enough: A randomized controlledtrial of the effects of HRQL assessment on quality oflife and satisfaction in oncology clinical practice.Psychooncology 16:1069-1079, 2007

32. Ruland CM, Holte HH, Røislien J, et al: Effectsof a computer-supported interactive tailored patientassessment tool on patient care, symptom distress,and patients’ need for symptom management sup-port: A randomized clinical trial. J Am Med InformAssoc 17:403-410, 2010

33. Sarna L: Effectiveness of structured nursingassessment of symptom distress in advanced lungcancer. Oncol Nurs Forum 25:1041-1048, 1998

34. Taenzer P, Bultz BD, Carlson LE, et al: Impactof computerized quality of life screening on physi-cian behaviour and patient satisfaction in lung can-cer outpatients. Psychooncology 9:203-213, 2000

35. Takeuchi EE, Keding A, Awad N, et al: Impactof patient-reported outcomes in oncology: A longi-tudinal analysis of patient-physician communication.J Clin Oncol 29:2910-2917, 2011

36. Thewes B, Butow P, Stuart-Harris R: Doesroutine psychological screening of newly diagnosedrural cancer patients lead to better patient out-comes? Results of a pilot study. Aust J Rural Health17:298-304, 2009

37. Trowbridge R, Dugan W, Jay SJ, et al: Deter-mining the effectiveness of a clinical-practice inter-vention in improving the control of pain inoutpatients with cancer. Acad Med 72:798-800,1997

38. Velikova G, Booth L, Smith AB, et al: Measur-ing quality of life in routine oncology practice im-proves communication and patient well-being: Arandomized controlled trial. J Clin Oncol 22:714-724,2004

39. Velikova G, Keding A, Harley C, et al: Patientsreport improvements in continuity of care when quality oflife assessments are used routinely in oncology practice:Secondary outcomes of a randomised controlled trial. EurJ Cancer 46:2381-2388, 2010

40. Cleeland CS, Wang XS, Shi Q, et al: Auto-mated symptom alerts reduce postoperative symp-tom severity after cancer surgery: A randomizedcontrolled clinical trial. J Clin Oncol 29:994-1000,2011

41. Kearney N, McCann L, Norrie J, et al: Evalua-tion of a mobile phone-based, advanced symptommanagement system (ASyMS) in the managementof chemotherapy-related toxicity. Support Care Can-cer 17:437-444, 2009

42. Kornblith AB, Dowell JM, Herndon JE II, et al:Telephone monitoring of distress in patients aged65 years or older with advanced stage cancer: Acancer and leukemia group B study. Cancer 107:2706-2714, 2006

43. Moher D, Hopewell S, Schulz KF, et al: CON-SORT 2010 explanation and elaboration: Updatedguidelines for reporting parallel group randomisedtrials. J Clin Epidemiol 63:e1-e37, 2010

44. Boyce MB, Browne JP: Does providing feed-back on patient-reported outcomes to healthcareprofessionals result in better outcomes for patients?A systematic review. Qual Life Res 22:2265-2278,2013

45. Devlin N, Appleby J: Getting the most out ofPROMs: Putting health outcomes at the heart of NHSdecision-making. London, United Kingdom, The King’sFund, 2010. https://www.kingsfund.org.uk/sites/files/kf/Getting-the-most-out-of-PROMs-Nancy-Devlin-John-Appleby-Kings-Fund-March-2010.pdf

■ ■ ■




https://www.kingsfund.org.uk/sites/files/kf/Getting-the-most-out-of-PROMs-Nancy-Devlin-John-Appleby-Kings-Fund-March-2010.pdf



Appendix

Table A1. Electronic Databases Searched and Search Terms Used

Electronic Databases Search Terms Used

Medline (1946 to May 2012) 1. exp controlled clinical trial/EMBASE (1974 to May 2012) 2. exp randomized controlled trial/CINAHL (inception to May 2012) 3. 1 OR 2PsycINFO (inception to May 2012) 4. exp neoplasms/OR cancer�.mp. OR neoplasm�.mp. OR carcinoma�.mp. OR oncol�.mp. OR

malignan�.mp. OR tumor�.mp. OR tumour�.mp. OR leukemia�.mp. OR leukaemia�.mp. ORsarcoma�.mp. OR lymphoma�.mp. OR melanoma�.mp. OR blastoma�.mp.

PBSC (inception to May 2012) 5. 3 AND 46. (patient reported outcomes OR patient reported outcome OR patient based outcome OR patient

reported outcome measure$).mp.7. inventory.ti. OR inventory.ab.8. instrument�.ti. OR instrument�.ab.9. measure�.ti.

10. self report�.ti. OR self report�.ab.11. 7 OR 8 OR 9 OR 1012. 6 OR 1113. 5 AND 1214. Remove duplicates from 1315. Limit 14 to abstracts16. Limit 15 to English language

NOTE: Search strategy as conducted in Ovid Medline.Abbreviations: ab, abstract; CINAHL, Cumulative Index to Nursing and Allied Health Literature; exp, term explosion; mp, free text search for a term; PBSC,

Psychology and Behavioral Sciences Collection; ti, title.

Kotronoulas et al

© 2014 by American Society of Clinical Oncology JOURNAL OF CLINICAL ONCOLOGY


Table A2. Intervention and Outcome Assessment PROMs Used in the 24 Studies Reviewed (n � 26 articles)

Author and PublicationYear Intervention PROM(s) Outcome Assessment PROM(s)�

Same Intervention/OutcomePROM(s)

Berry et al,17 2011 SDS Audio-recorded consultations NoEORTC QLQ-C30Pain scalePHQ-9SSS

Author-developed questionnaire regarding clinic visitduration; clinician evaluation of the intervention

Boyes et al,18 2006 Physical symptoms scales Physical symptoms scales Yes, plus additional PROMsHADS HADSSCNS-SF31 SCNS-SF31

Patient/clinician acceptability surveyBraeken et al,19 2011 SIPP Medical records

Intervention evaluation inventoriesNo

Carlson et al,20 2010 DT and problem list DT and problem list YesPSSCAN part C PSSCAN part C

Cleeland et al,40 2011 MDASI MDASI YesAuthor-developed form for patient evaluation of the

interventionDetmar et al,21 2002 EORTC QLQ-C30 Audio-recorded consultations No

COOPWONCAMedical recordsAuthor-developed fatigue scalePatient Satisfaction Questionnaire CPhysician satisfaction with communicationSF-36Patient/physician evaluation of the intervention

surveyGanz et al,22 2000 Daily diary symptom cards Daily diary symptom cards Yes, plus additional PROMs

CARES (sexual summary scale) CARES (sexual summary scale)RAND Vitality Scale

Girgis et al,23 2009 HADS HADS Yes, plus additional PROMsEORTC QLQ-C30 EORTC QLQ-C30SCNS-SF34 SCNS-SF34NA-ACP NA-ACP

Patient perceptions of improved communicationHilarius et al,24 2008 EORTC QLC-C30 Self-report communication questionnaire No

EORTC LC13 COOPEORTC BR23 WONCAEORTC CR38 Chart audit

PSQ-IISF-36FACT-L/C/BCSPatient/nurse evaluation of the intervention

questionnaireHoekstra et al,25 2006 The Symptom Monitor The Symptom Monitor YesKearney et al,41 2009 Author-developed symptom

questionnaire integrating theCTCAE grading system and theCSAS (electronic version)

Author-developed symptom questionnaireintegrating the CTCAE grading system and theCSAS (paper-based version)

Yes

Klinkhammer-Schalke etal,26 2012

EORTC QLC-C30 EORTC QLC-C30 Yes, plus additional PROMsEORTC BR23 EORTC BR23

Medical recordsKornblith et al,42 2006 HADS HADS Yes, plus additional PROMs

EORTC QLQ-C30 EORTC QLQ-C30MOS-SSS MOS-SSS

GDS-SFOARSQ, physical health subscaleUtilization of mental health and psychosocial

services scaleGSREPatient satisfaction with research program

(continued on following page)


www.jco.org © 2014 by American Society of Clinical Oncology


Table A2. Intervention and Outcome Assessment PROMs Used in the 24 Studies Reviewed (n � 26 articles) (continued)

Author and PublicationYear Intervention PROM(s) Outcome Assessment PROM(s)�

Same Intervention/OutcomePROM(s)

Maunsell et al,27 1996 GHQ-20 GHQ-20 Yes, plus additional PROMsSocial Support QuestionnaireLESLWMATPSIPerceptions of health and worries about healthNumber of visits to HPMedical records

McLachlan et al,28 2001 CNQ-SF CNQ-SF Yes, plus additional PROMsEORTC QLQ-C30 EORTC QLQ-C30BDI-SF BDI-SF

Patient satisfaction surveyMills et al,29 2009 EORTC QLQ-C30 FACT-L TOI subscale No

EORTC LC13 PQLIUtilization of diaryPatient/clinician communication checklistPatient satisfaction with care survey

Nicklasson et al,30 2013 EORTC QLQ-C30 Audio-recorded consultations NoEORTC LC13 Medical records

Rosenbloom et al,31

2007FACT-G FLIC No

Brief POMS-17PSQ-IIIClinical treatment changes survey

Ruland et al,32 2010 Choice ITPA Choice ITPA Yes, plus additional PROMsChart audit

Sarna,33 1998 SDS SDS YesTaenzer et al,34 2000 EORTC QLQ-C30 PDIS No

Exit interviewMedical record audit

Takeuchi et al,35 2011 EORTC QLQ-C30 Audio-recorded consultations NoHADS

Thewes et al,36 2009 DT Medical records NoSPHERE-Short SCNS-SF34

Satisfaction with intervention, Likert scalesTrowbridge et al,37

1997Pain inventories Pain inventories Yes, plus additional PROMs

PMIChart audit

Velikova et al,38 2004 EORTC QLQ-C30 Audio-recorded consultations NoHADS FACT-G

Physician use of QoL information checklistVelikova et al,39 2010 EORTC QLQ-C30 MCQ No

HADS Satisfaction with care, single-item scalesIntervention evaluation questionnaires

Abbreviations: BDI, Beck Depression Inventory; Brief POMS-17, Brief Profile of Mood States-17; BR-23, Breast Cancer 23 Module; CARES, Cancer RehabilitationEvaluation System; CNQ-SF, Cancer Needs Questionnaire–Short Form; COOP, Dartmouth Primary Care Cooperative Information Functional Health Assessment;CR-38, Colorectal Cancer 38 Module; CSAS, Chemotherapy Symptom Assessment Scale; CTCAE, Common Toxicity Criteria Adverse Events; DT, DistressThermometer; EORTC-LC13, European Organisation for Research and Treatment of Cancer–Lung Cancer Module 13; EORTC QLQ-C30, European Organisation forResearch and Treatment of Cancer–Core Quality of Life Questionnaire, version 3.0; FACT-G, Functional Assessment of Cancer Therapy–General; FACT-L/C/BCS,Functional Assessment of Cancer Therapy–Lung/Colorectal/Breast Cancer Subscale; FLIC, Functional Living Index Cancer; GDS-SF, Geriatric Depression Scale–ShortForm; GHQ, General Health Questionnaire; GSRE, Geriatric Schedule of Recent Experience; HADS, Hospital Anxiety and Depression Scale; HP, health professional;ITPA, interactive tailored patient assessments; LC-13, Lung Cancer 13 Module; LES, Life Experiences Survey; LWMAT, Lock-Wallace Marital Adjustment Test;MCQ, Medical Care Questionnaire; MDASI, MD Anderson Symptom Inventory; MOS-SSS, Medical Outcomes Study–Social Support Survey; NA-ACP, NeedsAssessment for Advanced Cancer Patients; OARSQ-Physical Health, Older American Resources and Services Questionnaire–Physical Health; PDIS, PatientSatisfaction Questionnaire; PHQ-9, Patient Health Questionnaire–9; PMI, Pain Management Index; PQLI, Palliative Care Quality of Life Index; PROM,patient-reported outcome measure; PSI, Psychiatric Symptom Index; PSQ-III/II, Medical Outcomes Study–Patient Satisfaction Questionnaire III/II; PSSCAN Part C,Psychological Screen for Cancer–Part C; QoL, quality of life; RAND, Research and Development; SCNS-SF31, Supportive Care Needs Survey–Short Form 31;SCNS-SF34, Supportive Care Needs Survey–Short Form 34; SDS, Symptom Distress Scale; SF-36, Medical Outcomes Study 36-Item Short Form Health Survey;SIPP, Screening Inventory of Psychosocial Problems; SPHERE-Short, Somatic and Psychological Health Report–Short form; SSS, Subject Significance Scale; TOI,Trial Outcome Index; WONCA, World Organization Project of National Colleges and Academics.

�If no specific PROM was used, method of assessment is reported instead.

Kotronoulas et al



Table A3. Evaluation of PROM Intervention Effects on Patient Outcomes

Outcome ES (d) 95% CI�† Effect Characterization‡

Menopausal symptom distress �1.18 �1.68 to �0.6722 �

PrevalenceAnxiety �0.07 �0.41 to 0.2723 �

Depression �0.15 �0.73 to 0.4323 �

Overall supportive care needs �0.20 �0.46 to 0.0623 �

0.5836 �

Need for helpPsychological needs �0.16 �0.73 to 0.4018 �

0.5036 �

Information needs �0.29 �0.86 to 0.2818 �

0.5336 �

Patient care and support �0.47 �1.05 to 0.1018 �

Physical and daily living �0.34 �0.91 to 0.2418 �

0.4636 �

Sexual functioning �0.49 �0.96 to �0.0222 �

QoLRole functioning �0.04 �0.26 to 0.1923 �

�0.12 �0.40 to 0.1621 �

Emotional/psychological functioning �0.18 �0.41 to 0.0523 �

�0.1131 �

�0.20 �0.48 to 0.0721 �

0.10 �0.25 to 0.4442 �

Cognitive functioning �0.05 �0.27 to 0.1823 �

Social functioning �0.01 �0.24 to 0.2223 �

�0.0431 �

�0.07 �0.35 to 0.2121 �

Physical functioning �0.16 �0.39 to 0.0123 �

�0.1231 �

�0.04 �0.32 to 0.2421 �

�0.20 �0.55 to 0.1542 �

Physical and functional well-being �0.41 �0.95 to 0.1429 �

Mental health �0.10 �0.38 to 0.1821 �

Vitality 0.08 �0.38 to 0.5422 �

�0.08 �0.36 to 0.2021 �

Bodily pain �0.07 �0.35 to 0.2121 �

Nausea �0.1631 �

Hardship owing to cancer �0.0531 �

Overall QoL �0.05 �0.28 to 0.1723 �

�0.1431 �

�0.59 �1.16 to �0.0129 �

�0.35 �0.70 to �0.00126 �

�0.04 �0.38 to 0.3142 �

SeverityFatigue �0.37 �0.77 to 0.0425 �

�0.25 �0.63 to 0.1241 �

Pain 0.04 �0.36 to 0.4425 �

Lack of appetite �0.04 �0.44 to 0.3625 �

Shortness of breath 0.05 �0.35 to 0.4525 �

Sore mouth/throat 0.32 �0.05 to 0.6941 �

Coughing �0.37 �0.77 to 0.0325 �

Sleeplessness �0.31 �0.71 to 0.0925 �

Hand-foot syndrome 0.42 0.05 to 0.7941 -Nausea �0.44 �0.84 to 0.0425 �

�0.18 �0.55 to 0.2041 �

Constipation 0.24 �0.16 to 0.6425 �

Diarrhea 0.0 �0.40 to 0.4025 �

0.06 �0.32 to 0.4341 �

Vomiting 0.33 �0.07 to 0.7325 �

0.01 �0.36 to 0.3841 �

Anxiety �0.09 �0.65 to 0.4818 �

�0.0520 �

�0.30 �0.65 to 0.0442 �





Table A3. Evaluation of PROM Intervention Effects on Patient Outcomes (continued)


Depression 0.08 �0.49 to 0.6418 �

�0.0120 �

�0.15 �0.49 to 0.2042 �

Psychological distress �0.09 �0.34 to 0.1627 �

�0.42 �0.76 to �0.0742 �

PrevalenceFatigue �0.07 �0.62 to 0.4725 �

�0.29 �0.60 to 0.0218 �

�0.2041 �

Pain �0.33 �0.78 to 0.1225 �

Lack of appetite �0.29 �0.74 to 0.1525 �

�0.19 �0.55 to 0.1818 �

Shortness of breath �0.06 �0.50 to 0.3825 �

Coughing 0.34 �0.11 to 0.7925 �

Sleeplessness �0.40 �0.85 to 0.0425 �

Nausea �0.10 �0.57 to 0.3725 �

�0.06 �0.79 to 0.6718 �

�0.1041 �

Constipation �0.73 �1.29 to �0.1725 �

�0.06 �1.60 to 1.4918 �

Diarrhea �0.32 �0.90 to 0.2725 �

�0.88 �2.10 to 0.3718 �

0.0141 �

Vomiting �0.98 �1.83 to �0.1325 �

�0.0541 �

Skin rash �0.06 �1.60 to 1.4918 �

Sore mouth 0.25 �0.58 to 1.0818 �

0.0641 �

Metallic taste �0.06 �1.17 to 1.0518 �

Hot flashes 0.75 �0.48 to 1.9818 �

Hand-foot syndrome 0.2341 �

Overall distress �0.1520 �

DistressVomiting 0.05 �0.32 to 0.4241 �

Nausea �0.15 �0.52 to 0.2241 �

Diarrhea 0.0 �0.37 to 0.3741 �

Hand-foot syndrome 0.35 �0.02 to 0.7241 �

Sore mouth/throat 0.33 �0.05 to 0.7041 �

Fatigue �0.31 �0.69 to 0.0641 �

Overall �0.0231 �

�0.1620 �

Health status �0.01 �0.34 to 0.3327 �

0.0 �0.34 to 0.3442

Worry about health �0.10 �0.39 to 0.2027 �

Working during assessment 0.01 �0.28 to 0.3027 �

Hours worked per week �0.05 �0.30 to 0.2027 �

Household activities performed �0.08 �0.33 to 0.1727 �

Engagement in social activities �0.23 �0.48 to 0.0227 �

Engagement in leisure activities 0.14 �0.11 to 0.3927 �

Engagement in physical activities �0.02 �0.27 to 0.2327 �

Marital satisfaction 0.0 �0.25 to 0.2527 �

NOTE. Negative ES denote more favorable outcomes (eg, less severity or better scores) for the intervention group, and vice versa. ES were not calculated forcontrolled trials that reported pre-intervention between-group differences in the outcome in question, or where no relevant data were available. Where data wereavailable, but no such baseline comparisons were performed/stated, baseline scores/percentages were compared using two-tailed independent sample t tests, thusensuring that postintervention scores were not a result of preintervention differences. When studies reported results at more than one time point, the final timepoint was used, thus ensuring independence of data; hence, each study contributed no more than one ES for a specific outcome.14 For studies with more than oneexperimental group, separate ES were calculated if different intervention PROMs were used; however, if the same intervention PROM was used, one ES wascalculated based on pooled experimental versus control effects. If a study indicated that the effect was nonsignificant but no statistics were provided, ES wasentered as zero.

Abbreviations: ES, effect sizes; PROM, patient-reported outcome measure; QoL, quality of life.�ES calculations were performed only in those studies for which enough data were available.†Where no 95% CIs are reported, not enough data were available to calculate them.‡Based on P values (P � .05) and direction; � favors the intervention group (P � .05); � favors the control group (P � .05); � represents P � .05.

Kotronoulas et al



Table A4. Evaluation of PROM Intervention Effects on Processes of Care


ActionEnrolled onto medical trial �0.15 �0.53 to 0.2227 �

Met with other survivors �0.14 �0.41 to 0.1427 �

Participated in patient-support group �0.02 �0.47 to 0.4327 �

Consulted treating oncologist �0.22 �0.52 to 0.0924 �

Consulted family physician �0.002 �0.33 to 0.3227 �

Consulted other physician �0.10 �0.38 to 0.1827 �

Had consultation for CAM therapies �0.18 �0.54 to 0.1927 �

Had psychiatric/psychological consultation �0.02 �0.44 to 0.4027 �

Sought help because of feeling depressed/sad �0.11 �0.41 to 0.1927 �

Had a confidant �0.27 �0.67 to 0.1327 �

Participated in relaxation activities �0.05 �0.43 to 0.3227 �

Made dietary changes 0.13 �0.14 to 0.4127 �

DiscussedNausea/vomiting �0.06 �0.26 to 0.1417 �

0.22 �0.08 to 0.5224 �

0.0238 �

�0.07 �0.41 to 0.2721 �

Appetite �0.06 �0.24 to 0.1317 �

�0.09 �0.41 to 0.2224 �

�0.4038 �

�0.34 �0.65 to �0.0330 �

0.06 �0.25 to 0.3721 �

Insomnia/sleep problems �0.05 �0.23 to 0.1317 �

�0.66 �1.00 to �0.3224 �

�0.6438 �

�0.13 �0.50 to 0.2421 �

Pain 0.02 �0.16 to 0.1917 �

�0.10 �0.39 to 0.2024 �

�0.0138 �

�0.05 �0.36 to 0.2530 �

�0.30 �0.62 to 0.0321 �

Fatigue 0.0 �0.19 to 0.1917 �

�0.13 �0.43 to 0.1724 �

�0.3438 �

�0.06 �0.36 to 0.2530 �

�0.38 �0.69 to �0.0721 �

Bowel pattern 0.14 �0.05 to 0.3317 �

Constipation �0.40 �0.72 to �0.0824 �

Diarrhea �0.67 �1.04 to �0.3024 �

Concentration �0.29 �0.64 to 0.0717 �

Appearance 0.19 �0.07 to 0.4517 �

Impact on sex �0.58 �0.99 to �0.1717 �

Breathing/dyspnea 0.01 �0.18 to 0.1917 �

�0.77 �1.22 to �0.3324 �

�0.1538 �

�0.18 �0.48 to 0.1330 �

�0.40 �0.82 to 0.0221 �

Outlook �0.05 �0.24 to 0.1517 �

Cough �0.05 �0.24 to 0.1417 �

Fever/chills �0.03 �0.21 to 0.1517 �

Depression �0.12 �0.36 to 0.1317 �

Suicidal ideation �0.26 �0.89 to 0.3617 �

Symptoms of illness �0.07 �0.38 to 0.2330 �

0.02 �0.52 to 0.5629 �

Physical functioning 0.03 �0.15 to 0.2117 �

0.26 �0.10 to 0.6324 �

�0.2138 �

�0.18 �0.48 to 0.1330 �

�0.98 �1.31 to �0.6421 �

�0.05 �0.26 to 0.1719 �





Table A4. Evaluation of PROM Intervention Effects on Processes of Care (continued)


Emotional functioning �0.11 �0.28 to 0.0717 �

0.05 �0.28 to 0.3724 �

�0.2438 �

�0.44 �0.75 to �0.1330 �

�0.17 �0.48 to 0.1421 �

0.26 �0.29 to 0.8129 �

�0.19 �0.38 to �0.0119 �

Social functioning �0.14 �0.37 to 0.0817 �

�0.18 �0.59 to 0.2324 �

0.1938 �

�0.21 �0.51 to 0.1030 �

0.05 �0.49 to 0.6229 �

�0.49 �0.93 to �0.0421 �

�0.16 �0.38 to 0.0619 �

Cognitive functioning �0.08 �0.35 to 0.1817 �

�0.66 �1.19 to �0.1224 �

�0.3338 �

0.0 �0.31 to 0.3130 �

�0.36 �0.97 to 0.2521 �

Daily functioning 0.14 �0.22 to 0.5024 �

0.38 �0.19 to 0.9429 �

Role functioning 0.01 �0.18 to 0.2017 �

�0.1538 �

0.33 �0.23 to 0.9029 �

0.70 0.37 to 1.0321 -Sexual problems 0.06 �0.16 to 0.2819 �

Impact on family relationships 0.30 �0.25 to 0.8529 �

Existential issues 0.0 �0.31 to 0.3130 �

Financial issues 0.10 �0.20 to 0.4130 �

0.02 �0.53 to 0.5829 �

Medical/technical issues/effects of treatment 0.27 �0.04 to 0.5730 �

0.23 �0.33 to 0.7829 �

Overall condition 0.37 �0.20 to 0.9529 �

Global QoL �0.01 �0.24 to 0.2117 �

�0.45 �0.76 to �0.1430 �

0.10 �0.21 to 0.4130 �

No. of concerns/symptoms discussed during consultations �1.09 �1.67 to �0.5234 �

�0.4138 �

�0.38 �0.66 to �0.1021 �

No. of concerns/issues charted on patient records by nurses �0.54 �0.81 to �0.2724 �

�0.6832 �

No. of concerns/issues charted on patient records by physicians �0.3332 �

No. of concerns/issues charted on patient records by healthprofessionals, mixed sample �0.49 �1.04 to 0.0534 �

Average duration of contact 0.18 �0.07 to 0.4327 �

�0.08 �0.24 to 0.0917 �

0.12 �0.13 to 0.3728 �

0.0938 �

0.03 �0.27 to 0.3330 �

0.09 �0.19 to 0.3721 �

Satisfaction with nursing care �0.56 �1.40 to 0.2828 �

Satisfaction with medical care �0.16 �1.16 to 0.8428 �

Satisfaction with information received �0.50 �1.12 to 0.1228 �

0.18 �0.36 to 0.7234 �

0.03 �0.53 to 0.6029 �

Satisfaction with support/rapport/communication 0.0 �0.54 to 0.5434 �

�0.0731 �

�0.04 �0.61 to 0.5329 �

�0.37 �0.65 to �0.0921 �

0.13 �0.05 to 0.3119 �


Kotronoulas et al



Table A4. Evaluation of PROM Intervention Effects on Processes of Care (continued)


Satisfaction with help received about important problems 0.69 0.20 to 1.1742 -Satisfaction with involvement in decision-making 0.14 �0.42 to 0.7129 �

Satisfaction with HPs addressing patient needs �0.35 �0.90 to 0.1934 �

0.13 �0.44 to 0.6929 �

Overall satisfaction with care �0.39 �1.92 to 1.1528 �

�0.0839 �

0.3331 �

0.13 �0.44 to 0.6929 �

Overall satisfaction with intervention �0.52 �1.03 to �0.0142 �

Intervention acceptability, comfortable with using thesystem �0.49 �0.94 to �0.0440 �

Intervention acceptability, system easy to use �0.59 �0.14 to �1.0540 �

HP satisfaction with clinical encounter 0.021 �

HP actionNo. of actions taken/medical decisions made per patient �0.40 �0.94 to 0.1534 �

0.1638 �

0.0231 �

�0.32 �0.62 to �0.0230 �

Referred to psychosocial care or other provider 0.08 �0.27 to 0.4219 �

�0.31 �0.87 to 0.2636 �

�0.01 �0.31 to 0.2824 �

0.11 �0.22 to 0.4326 �

�0.3220 �

0.04 �0.18 to 0.2619 �

Prescription of medication 0.26 �0.07 to 0.6024 �

�0.41 �0.72 to �0.0937 �

Ordering tests �0.11 �0.45 to 0.2224 �

Changing/stopping chemotherapy �0.05 �0.38 to 0.2724 �

Offering counseling on managing health problems �0.26 �0.65 to 0.1421 �

HP awareness of patient outcomesPhysical �0.13 �0.40 to 0.1424 �

�0.21 �0.69 to 0.2721 �

Feelings �0.16 �0.43 to 0.1124 �

�0.13 �0.66 to 0.3921 �

Daily activities �0.28 �0.55 to �0.0124 �

0.09 �0.41 to 0.5921 �

Social activities �0.09 �0.35 to 0.1824 �

�0.50 �1.05 to 0.0521 �

Overall health �0.20 �0.47 to 0.0724 �

0.19 �0.27 to 0.6421 �

Pain �0.54 �0.82 to �0.2724 �

0.20 �0.34 to 0.7421 �

Fatigue �0.15 �0.41 to 0.1224 �

�0.18 �0.58 to 0.2321 �

QoL �0.27 �0.54 to 0.024 �

Prevalence of patients undertreated for pain �0.07 �0.33 to 0.1837 �

NOTE. Negative ES denote more favorable outcomes (ie, more frequent discussion or better communication) for the intervention group and vice versa. ES werenot calculated for controlled trials that reported preintervention between-group differences in the outcome in question or where no relevant data were available.Where data were available but no such baseline comparisons were performed/stated, baseline scores/percentages were compared using two-tailed independentsample t tests, thus ensuring that postintervention scores were not because of preintervention differences. When studies reported results at more than one timepoint, the final time point was used, thus ensuring independence of data. Hence, each study contributed no more than one ES for a specific outcome.14 For studieswith more than one experimental group, separate ES were calculated if different intervention PROMs were used; however, if the same intervention PROM wasused, one ES was calculated based on pooled experimental versus control effects. If a study indicated that the effect was not significant but no statistics wereprovided, ES was entered as zero.

Abbreviations: CAM, complementary/alternative medicine; ES, effect sizes; HP, health professional; PROM, patient-reported outcome measure; QoL, quality of life.�ES calculations were performed only in those studies for which enough data were available.†Where no 95% CIs are reported, not enough data were available to calculate them.‡Based on P value (P � .05) and direction; � favors the intervention group (P � .05); � favors the control group (P � .05); � represents P � .05.




Table A5. Evaluation of PROM Intervention Effects on Health Service Outcomes

Outcome ES (d) 95% CI� Effect Characterization†

Patient use of psychological referrals �0.10 �1.02 to 0.8222 �

Self-referrals �0.20 �0.44 to 0.0420 �

Patient contacts with health professional �0.85 �1.10 to �0.5927 �

�0.15 �0.45 to 0.1530 �

Patient use of mental health services 0.18 �0.45 to 0.8242 �

NOTE. Negative effect sizes denote more favorable outcomes (eg, more frequent use of service or more contacts) for the intervention group and vice versa. ESwere not calculated for controlled trials that reported preintervention between-group differences in the outcome in question or where no relevant data were available.Where data were available but no such baseline comparisons were performed or stated, baseline scores/percentages were compared using two-tailed independentsample t tests, thus ensuring that postintervention scores were not because of preintervention differences. When studies reported results at more than one timepoint, the final time point was used, thus ensuring independence of data. Hence, each study contributed no more than one ES for a specific outcome.14 For studieswith more than one experimental group, separate ES were calculated if different intervention PROMs were used; however, if the same intervention PROM wasused, one ES was calculated based on pooled experimental versus control effects. If a study indicated that the effect was nonsignificant but no statistics wereprovided, ES was entered as zero.

Abbreviations: ES, effect size; PROM, patient-reported outcome measure.�ES calculations were performed only in those studies for which enough data were available.†Based on P value (P � .05) and direction; � (P � .05 favors intervention group); � P � .05 favors control group); � (P � .05).

Kotronoulas et al



What Is the Value of the Routine Use of Patient-Reported ...

Documents

Transcript of What Is the Value of the Routine Use of Patient-Reported ...