PSYCHOSOCIAL INTERVENTIONS
FOR ADOLESCENTS AND YOUNG
ADULTS WITH CANCER
CELESTE R. PHILLIPS AND LORIE L. DAVIS
C
ancer is 2.9 times more likely to occur survival rates of young children, AYA with cancer
in adolescents and young adults (AYA) have not seen the same improvement.2 Over the
ages 15 to 29 than in younger chil- past decade, there has been growing international
dren.1 Although medical advance- attention on addressing the specific needs of AYA
ments in pediatric oncology have improved the with cancer.3-6 AYA not only have unique medical
needs, but they also have many unique emotional,
social, spiritual, and physical needs. Recent
Celeste R. Phillips, PhD, RN, CPONÒ: Assistant research suggests that AYA are inadequately
Professor, Indiana University School of Nursing, Indi- served by current support services.7-9
anapolis, IN. Lorie L. Davis, MSN, RN, OCNÒ: Doctoral Despite the growing international attention, there
Student, Indiana University School of Nursing, Indian- are few psychosocial interventions designed specif-
apolis, IN.
ically for AYA with cancer. In 2009, Seitz and col-
Address correspondence to Celeste R. Phillips, PhD,
leagues,10 in a systematic review of peer-reviewed
RN, CPONÒ, 1111 Middle Dr., NU 425E, Indiana Uni-
versity School of Nursing, Indianapolis, IN 46202. articles reporting on psychological or psychosocial
e-mail: cephilli@iu.edu interventions for adolescents with cancer between
Ó 2015 Elsevier Inc. All rights reserved. 1988 and 2007, found only four such studies. These
0749-2081/3103-$36.00/0. studies had small sample sizes (n ¼ 14 to 78), and
http://dx.doi.org/10.1016/j.soncn.2015.05.004 only one, a small pilot study (n ¼ 21), reported
UPDATED PSYCHOSOCIAL INTERVENTIONS 243
significant improvement compared with a waitlist characteristics including age, gender, and ethnicity;
control group. We sought to update findings from type of study design; and theoretical framework.
the previously published review by conducting a Table 2 includes an overview of the interventions
systematic review of psychosocial interventions de- with abstracted data addressing: behavior(s)
signed for AYA with cancer published since 2007 encouraged: mode of delivery (ie, route and inter-
and by summarizing and evaluating these studies. vener, if applicable): timing of delivery: duration:
a description of the intervention: and a descrip-
tion of the control group. Table 3 describes the
METHODS methodological quality of the studies. Articles
were critiqued according to the following salient
Search Strategy features: randomization and processes descri-
To identify relevant studies addressing interven- bed (yes/no or not applicable for randomized,
tions for AYA with cancer, the following databases randomization processes described, and blinding),
were searched: PubMed, Ovid, and PsycINFO. In power analysis (yes/no reported), length of follow-
addition to the database search, reference lists of up, attrition, and the number of participants in
all relevant studies and review articles were each arm that completed all measures. Table 4 dis-
scanned for further references that met the inclu- plays the outcome variables of interest, measures
sion criteria. used, and significant findings.
The inclusion criteria were: (1) full-length, peer-
reviewed articles published between January 2008
and December 2014; (2) a study sample of AYA can- RESULTS
cer patients between the ages of 10 and 29; (3) En-
glish language; (4) uncontrolled or controlled The search produced only five studies that met
trials; and (5) measurement of psychosocial our inclusion criteria.12-16 One study included a
outcome variables. Exclusion criteria were: case sample of AYA who ranged from 9 to 20 years of
studies, editorials, abstracts, dissertations, and age,16 which was slightly outside our pre-
studies focused on adolescent survivors of child- specified age range of 10 to 29. We decided to
hood cancers. Although the National Cancer Insti- include this article because the majority of the
tute defines AYA as being between 15 and 39, for sample (78%) met this criterion (L. Wu, personal
this review we narrowed the age range to 15 to communication, March 2015).
29 years. This reflects the age range most commonly As shown in Table 1, three of the studies were
considered the AYA cancer population by the conducted in the United States12,14,15 and one in
oncology community and also narrows the develop- Taiwan.16 One international study was conducted
mental span for psychosocial interventions.11 in the United States, Canada, and Australia.13
Reference lists and abstracts for identified arti- Four studies were conducted with AYA,12,13,15,16
cles were scanned for their relevance. Using the and one study focused on the adolescent/family
key words: adolescents, young adults, cancer, dyad.14 Ages of AYA ranged from 9 to 29. Three
and interventions, a total of 242 articles were iden- studies included fewer than 100 partici-
tified; however, 233 were eliminated because they pants.12,14,16 The average age of research partici-
did not meet the inclusion criteria. Full copies of pants was not explicitly reported in one study13;
the nine articles that seemed to meet the inclusion instead, the percentage of participants in stratified
criteria were obtained. After reading these nine, age groups was reported. In the four other studies,
four were excluded because, for example, they mean age ranged from 13.2 to 17.3 years.12,14-16
used study samples of adolescent survivors of child- Gender was reported in all studies and females
hood cancer or gave descriptions of the interven- comprised from 31% to 45% of the samples.
tion protocol but did not describe results. Ethnicity was reported in four studies, and in
those studies from 50% to 75% of the samples
Data Organization were White.12-15 All studies were randomized
Data from the remaining five eligible articles were controlled trials; however, three were pilot
abstracted and organized into four separate studies.12,14,16 Theoretical frameworks were re-
tables by one person and verified by a second ported in only three of the reviewed studies.13,15,16
person. Table 1 focuses on: year published; As shown in Table 2, behavioral modification for
country of origin; population; sample size; sample active/effective coping, self-efficacy, and control
244
C.R. PHILLIPS AND L.L. DAVIS
TABLE 1.
Sample Demographics
Sample Characteristics
Gender Ethnicity Theoretical
Study Country Population N Age* (yrs) (% female) (% white) Design Framework
Jones (2010)12 US Adolescents (aged 12-18) 65 14.8 36.9 75.3 RCT/pilot Not specified
with solid tumors
Kato (2008)13 US/Canada/ Adolescents/young adults 371 13-14, 35.3%; 32.3 56.6 RCT Self-regulation Model of
Australia (aged 13-29) with 15-16, 30.7%; Health and Illness; Social
malignancies including 17-18, 21.3%; Cognitive Theory;
acute leukemia, 19-29, 12.7% Learning Theory
lymphoma, and sarcoma
Lyon (2014)14 US Adolescent (aged 30 dyads 16.3 40 50 RCT/pilot Not specified
14-20)/family dyads
with a cancer diagnosis
Robb (2014)15 US Adolescents/young adults 113 17.3 42.5 58.4 RCT Haase’s Resilience in
(aged 11-24) undergoing Illness Model (RIM);
stem cell transplant Robb’s Contextual
Support Model of Music
Therapy (CSM-MT)
Wu (2013)16 Taiwan Children/adolescents/ 58 14.1 (con) 44.8 (con.) Not RCT/pilot Lazarus’ Transactional
young adults with 13.2 (exp) 31 (exp.) reported theory
cancer (aged 9-20)
245
(Continued )
246 C.R. PHILLIPS AND L.L. DAVIS
Treatment as
ment preferences.14 Face-to-face contact as the
mode of delivery for the intervention was used in
usual
three studies.14-16 An interactive, asynchronous
CD-ROM or video game play was used in two
directly targeted Resilience
3 weeks; components
Brief Description
of Intervention
assistant
DISCUSSION
Wu (2013)16
Study)
Intervention: n ¼ 17 dyads
with the last review published10 only nine inter-
Control: n ¼ 13 dyads
Intervention: n ¼ 164
vention studies have been conducted with AYA
Intervention: n ¼ 35
Intervention: n ¼ 37
Intervention: n ¼ 29
with cancer since 1988 and only two have been
Control: n ¼ 140
Control: n ¼ 30
Control: n ¼ 30
Control: n ¼ 29
large randomized controlled trial studies,13,15
Furthermore, these few studies generally had
small samples and limited results, which serves
as a call to action for more intervention develop-
ment and evaluation for this underserved cancer
population.
17% attrition (intervention)
21% attrition (control)
Strengths
Attrition
7% attrition
T1: Baseline
T1: Baseline
T1: Baseline
T1: Baseline
iii. Yes
ii. Yes
ii. Yes
i. Yes
i. Yes
i. Yes
i. Yes
i. No
Robb (2014)15
Wu (2013)16
TABLE 4.
Outcome Variables and Significant Findings
Outcome Variables
Study of Interest Measure Significant Findings
12
Jones (2010) 1. Control Multidimensional Health Locus of Adolescents who received the
Control Scale Form B (MHLC-B) CD-ROM were significantly more
2. QOL Pediatric Oncology QOL Scale likely to increase their internal locus
(POQOLS) of control scores (P ¼ 0.016);
3. Coping Style KIDCOPE however, no significant differences
4. Self-efficacy Questionnaire developed for study were observed in regard to QOL,
5. Cancer knowledge Questionnaire developed for study coping, self-efficacy, or knowledge
6. Acceptability, use, Questionnaire developed for study
and satisfaction
Kato (2008)13 1. Treatment adherence Medication Adherence Scale (MAS); Self-reported treatment adherence did
Chronic Disease Compliance not differ significantly between
Instrument (CDCI) groups as measured by the MAS
2. Antibiotic adherence MEMS-cap monitoring; 6-MP blood and CDCI (group time interaction,
assays P ¼ 0.503; P ¼ 0.78, respectively);
3. Self-efficacy Questionnaire developed for study MEMS-cap monitoring showed a
4. Knowledge Questionnaire developed for study 16% increase in adherence in the
5. QOL Pediatric QOL– Generic Core Scale intervention group; A significantly
Version 4.0; Functional Assessment greater increase in cancer-related
of Cancer Therapy- General knowledge and cancer-specific
6. Stress Perceived Stress Scale self-efficacy was shown over time in
7. Control Multidimensional Health Locus of the intervention group (group time
Control Scale Form C interaction of P ¼ 0.035 and
P ¼ 0.011, respectively); The
intervention group did not
demonstrate significant
group time interactions for the
measures of adherence, stress,
control, or QOL
Lyon (2014)14 1. Anxiety Beck Anxiety Inventory Adolescents’ anxiety decreased
2. Depression Beck Depression Inventory significantly from baseline to
3. Pediatric QOL Pediatric QOL Inventory 4.0 Generic 3 months post-intervention in both
Core Scales groups (b ¼ 5.6; P ¼ 0.0212);
4. Spiritual Well-being Spiritual Well-Being Scale of the Total spirituality scores (b ¼ 8.1;
Functional Assessment of Chronic P ¼ 0.0296) were significantly
Illness Therapy- Version 4 higher among the intervention
5. Satisfaction with intervention Satisfaction Questionnaire versus control group. 100%
6. Feasibility of intervention % eligible and enrolled, % attendance attendance at all 3 sessions, 93%
at sessions, % retention at follow- retention at 3 months post-
up, and % data completeness at intervention, 100% of families rated
follow-up intervention as worthwhile, and
adolescents’ rating as worthwhile
increased over time (65% to 82%).
Robb (2014)15 1. Illness-related distress McCorkle Symptom Distress Scale; At T2 the intervention group reported
Mishel Uncertainty in Illness Scale significantly better courageous
2. Coping-defensive Jalowiec Coping Scale-Revised: coping (effect size, 0.505;
Emotive & Evasive Subscales P ¼ 0.030); At T3 the intervention
3. Spiritual perspective Reed Spiritual Perspective Scale group reported significantly better
4. Social integration Perceived Social Support-Health Care social integration (effect size, 0.543;
Providers; Perceived Social P ¼ 0.028) and family environment
Support-Friends; Perceived Social (effect size, 0.663; P ¼ 0.008);
Support-Family Moderate non-significant effect
5. Family environment Family Adaptability/Cohesion Scale II; sizes for spiritual perspective (effect
Parent-Adolescent Communication size, 0.450: P ¼ 0.071) and
Scale; Family Strengths Scale
(Continued )
UPDATED PSYCHOSOCIAL INTERVENTIONS 249
TABLE 4.
(Continued)
Outcome Variables
Study of Interest Measure Significant Findings
elements to support self-efficacy, control, or au- colleagues16 also encouraged the AYA to self-
tonomy, which corresponds to fostering healthy reflect on their treatment-related stressors,
adolescent development.18 discuss current coping styles, and then role-play
new coping techniques with a facilitator. Robb
Limitations and colleagues’ study was the only one that
Several limitations were also identified. First, fe- demonstrated significant results on coping.
males were somewhat underrepresented in all of The fourth limitation was that all but one
the studies. This is an important consideration study13 was underpowered (ie, #80%), suggesting
for future intervention studies because female that researchers need to enhance their AYA
AYA have been identified as having greater recruitment efforts and monitor the recruitment
distress than male AYA.19,20 Second, few age- process closely over the length of the grant period.
appropriate theories were used to guide the devel- The fifth limitation was that the length of post-
opment or evaluation of these studies. Of the three intervention follow-up tended to be fairly short
studies that used a theoretical framework, only (1 to 3 months). Therefore, it is unclear whether
Robb and colleagues15 used a theory that was or not these interventions had a long-term impact
age-appropriate. The other theories have most on the AYA. The sixth limitation was the lack of
commonly been evaluated with adult populations consistency in the outcome instruments used,
and may not hold true for the adolescent experi- which made it difficult to compare results across
ence. This limitation is likely caused by the lack the studies. Lastly, most of the studies only used
of well-developed and evaluated theories devel- self-report measures to evaluate the intervention.
oped or tested for the adolescent population, In conclusion, this update review indicates
particularly with cancer. Third, only two studies that the development and evaluation of age-
focused specifically on actively involving the appropriate psychosocial interventions for AYA
AYA in the development of effective coping skills with cancer is still in its infancy. However, we
through self-reflection and communication of are starting to finally see some larger randomized,
their cancer experience. Robb and colleagues15 controlled trials with promising results that incor-
provided a structured creativity opportunity for porate innovative technology, which is age-
AYA to self-reflect, identify, and express what appropriate and considered favorably by AYA.
was important to them through the development Future research should focus on conducting
of their own personalized music video. Wu and more multi-site studies to obtain larger sample
250 C.R. PHILLIPS AND L.L. DAVIS
sizes. Researchers will also need to overestimate IMPLICATIONS FOR NURSING PRACTICE
their targeted sample size because of the often se-
vere and complex nature of cancer to obtain Although two of the reviewed interventions13,15
adequate power to detect differences. Secondly, are ready for additional transitional research,
research should continue to develop interventions none are ready to be immediately translated into
based on the specific needs of AYA and to obtain clinical practice at this time. However, what can
their feedback in the development and/or be taken from this review is that it is extremely
improvement of these interventions. Additionally, important for nurses to evaluate and consider
as more intervention studies are conducted inter- the needs of AYA. Incorporating creative ways
nationally it will be important to evaluate cultural for the AYA to express their needs and concerns
differences so that efforts are made to support and to self-reflect on their experiences seems to
these potential differences. Lastly, consistency in be critically important and may be a way for
instruments to evaluate outcomes will help future nurses to help AYA cope positively with the cancer
comparison analyses. experience.
REFERENCES
1. Bleyer A, Viny A, Barr R. Cancer in 15- to 29-year-olds by 11. Bleyer A. The adolescent and young adult gap in cancer
primary site. Oncologist 2006;11:590-601. care and outcome. Curr Probl Pediatr Adolesc Health Care
2. Bleyer A, Budd T, Montello M. Adolescents and young 2005;35:182-217.
adults with cancer: The scope of the problem and criticality of 12. Jones JK, Kamani SA, Bush PJ, et al. Development and
clinical trials. Cancer 2006;107:1645-1655. Evaluation of an educational interactive CD-ROM for teens
3. Australian Institute of Health and Welfare (AIHW). Cancer with cancer. Pediatr Blood Cancer 2010;55:512-519.
in adolescents and young adults in Australia. 2011. Available at: 13. Kato PM, Cole SW, Bradlyn AS, Pollock BH. A video game
http://www.aihw.gov.au/publication-detail/?id¼10737420603. improves behavioral outcomes in adolescents and young adults
Accessed March 25, 2015. with cancer: a randomized trial. Pediatrics 2008;122:e305-
4. Adolescent and Young Adult Oncology Progress Review e317.
Group. Closing the gap: research and care imperatives for adoles- 14. Lyon ME, Jacobs S, Briggs L, Cheng YI, Wang J. A longi-
cents and young adults with cancer. 2006. Available at: http:// tudinal, randomized, controlled trial of advance care planning
planning.cancer.gov/library/AYAO_PRG_Report_2006_FINAL. for teens with cancer: Anxiety, depression, quality of life,
pdf. Accessed March 25, 2015. advance directives, spirituality. J Adolesc Health 2014;54:710-
5. Institute of Medicine (IOM). Identifying and addressing the 717.
needs of adolescents and young adults with cancer- Workshop 15. Robb SL, Burns DS, Stegenga KA, et al. Randomized clin-
summary. 2013. Available at: http://iom.edu/Reports/2013/ ical trial of therapeutic music video intervention for resilience
Identifying-and-Addressing-the-Needs-of-Adolescents-and-Young- outcomes in adolescents/young adults undergoing hematopoi-
Adults-with-Cancer.aspx. Accessed March 25, 2015. etic stem cell transplant. Cancer 2014;120:909-917.
6. Wilkins KL, D’Agostino N, Penney AM, et al. Supporting ad- 16. Wu LM, Chiou SS, Sheen JM, et al. Evaluating the
olescents and young adults with cancer through transitions: Po- acceptability and efficacy of a psycho-educational interven-
sition statement from the Canadian Task Force on Adolescents tion for coping and symptom management by children with
and Young Adults with cancer. J Pediatr Hematol Oncol cancer: a randomized controlled study. J Adv Nurs
2014;36:545-551. 2013;70:1653-1662.
7. Zebrack B, Hamilton R, Smith AW. Psychosocial outcomes 17. Zebrack B, Isaacson S. Psychosocial care of adolescent
and service use among young adults with cancer. Semin Oncol and young adult patients with cancer and survivors. J Clin On-
2009;36:468-477. col 2012;30:1221-1226.
8. Dyson GL, Thompson K, Palmer S, et al. The relationship 18. Austin S, Guay G, Senecal C, et al. Longitudinal testing of
between unmet needs and distress amongst young people with dietary self-care motivational model in adolescents with dia-
cancer. Support Care Cancer 2012;20:75-85. betes. J Pyschosom Res 2013;75:153-159.
9. Keegan TH, Lichtensztajn DY, Kato I, et al. Unmet adoles- 19. Wesley KM, Zelikovsky N, Schwartz LA. Physical symp-
cent and young adult cancer survivors information and service toms, perceived social support, and affect in adolescents with
needs: a population-based cancer registry study. J Cancer Sur- cancer. J Psychosoc Oncol 2013;31:451-467.
viv 2012;6:239-250. 20. Hinds PS, Nuss SL, Ruccione KS, et al. PROMIS pediatric
10. Seitz DC, Besier T, Goldbeck L. Psychosocial interven- measures in pediatric oncology: valid and clinically feasible in-
tions for adolescent cancer patients: a systematic review of dicators of patient-reported outcomes. Pediatr Blood Cancer
the literature. Pyschooncology 2009;18:683-690. 2013;60:402-408.