ORIGINAL CONTRIBUTION

Health Status of Adult Long-term

Survivors of Childhood Cancer
A Report From the Childhood Cancer Survivor Study
Melissa M. Hudson, MD Context Adult survivors of childhood cancer are at risk for medical and psychosocial
Ann C. Mertens, PhD sequelae that may adversely affect their health status.
Yutaka Yasui, PhD Objectives To compare the health status of adult survivors of childhood cancer and
siblings and to identify factors associated with adverse outcomes.
Wendy Hobbie, CRNP
Design, Setting, and Participants Health status was assessed in 9535 adult par-
Hegang Chen, PhD ticipants of the Childhood Cancer Survivor Study, a cohort of long-term survivors of
James G. Gurney, PhD childhood cancer who were diagnosed between 1970 and 1986. A randomly selected
cohort of the survivors’ siblings (n=2916) served as a comparison group.
Mark Yeazel, MD
Main Outcome Measures Six health status domains were assessed: general health,
Christopher J. Recklitis, PhD, MPH mental health, functional status, activity limitations, cancer-related pain, and cancer-
Neyssa Marina, MD related anxiety/fears. The first 4 domains were assessed in the control group.
Leslie R. Robison, PhD Results Survivors were significantly more likely to report adverse general health (odds
ratio [OR], 2.5; 95% confidence interval [CI], 2.1-3.0; P⬍.001), mental health (OR, 1.8;
Kevin C. Oeffinger, MD 95% CI, 1.6-2.1; P⬍.001), activity limitations (OR, 2.7; 95% CI, 2.3-3.3; P⬍.001), and
for the Childhood Cancer Survivor functional impairment (OR, 5.2; 95% CI, 4.1-6.6; P⬍.001), compared with siblings. Forty-
Study Investigators four percent of survivors reported at least 1 adversely affected health status domain. So-
ciodemographic factors associated with reporting at least 1 adverse health status domain

T
HE DEVELOPMENT OF CURA - included being female (OR, 1.4; 95% CI, 1.3-1.6; P⬍.001), lower level of educational
tive therapy for most pediat- attainment (OR, 2.0; 95% CI, 1.8-2.2; P⬍.001), and annual income less than $20000
ric malignancies provides the (OR, 1.8; 95% CI, 1.6-2.1; P⬍.001). Relative to those survivors with childhood leuke-
opportunity to investigate the mia, an increased risk was observed for at least 1 adverse health status domain among
impact of cancer treatment on health those with bone tumors (OR, 2.1; 95% CI, 1.8-2.5; P⬍.001), central nervous system tu-
status of large numbers of long-term mors (OR, 1.7; 95% CI, 1.5-2.0; P⬍.001), and sarcomas (OR, 1.2; 95% CI, 1.1-1.5; P=.01).
survivors. Numerous studies have es- Conclusion Clinicians caring for adult survivors of childhood cancer should be aware
tablished that cancer and its treatment of the substantial risk for adverse health status, especially among females, those with
predispose to late morbidity and in- low educational attainment, and those with low household incomes.
crease the risk of early mortality in long- JAMA. 2003;290:1583-1592 www.jama.com
term childhood cancer survivors.1-8 In
general, children with aggressive tu- and second malignancy. The fre- lated with sex, age at diagnosis, and
mor histologies have required more in- quency and severity of many of the cumulative dose-exposures of specific
tensive treatment, which predispose common sequelae have been corre- treatment modalities.9-12
them to heightened risks of physical
Author Affiliations: Department of Hematology Oncol- Institute, Boston, Mass (Dr Recklitis); Department of
morbidity. Treatment-related late se- ogy, St Jude Children’s Research Hospital and the Uni- Pediatrics, Stanford University Medical Center, Stan-
quelae with significant potential im- versity of Tennessee College of Medicine, Memphis (Dr ford, Calif (Dr Marina); Department of Family Practice
Hudson); Departments of Pediatrics (Drs Mertens, Gur- and Community Medicine, University of Texas South-
pact on physical functioning include ney, and Robison), Family Practice (Dr Yeazel), and Bio- western Medical Center at Dallas (Dr Oeffinger).
neurocognitive dysfunction, cardio- statistics (Dr Chen), University of Minnesota, Minne- The CCSS investigators are listed at the end of the
apolis; Cancer Prevention Research Program, Fred article.
pulmonary toxicity, endocrinopathy, Hutchinson Cancer Research Center, Seattle, Wash (Dr Corresponding Author and Reprints: Melissa M. Hud-
Yasui); Department of Oncology, Children’s Hospital son, MD, St Jude Children’s Research Hospital, 332
For editorial comment see p 1641. of Philadelphia, Philadelphia, Pa (Ms Hobbie); Depart- N Lauderdale, Memphis, TN 38105 (e-mail: melissa
ment of Pediatric Oncology, Dana Farber Cancer [email protected]).

©2003 American Medical Association. All rights reserved. (Reprinted) JAMA, September 24, 2003—Vol 290, No. 12 1583

Downloaded From: http://jama.jamanetwork.com/ by a Nanyang Technological University User on 05/19/2015

 HEALTH STATUS OF ADULT LONG-TERM CHILDHOOD CANCER SURVIVORS

In contrast, the impact of cancer and years after treatment for cancer, leuke- initial treatment, treatment for re-
its therapy on the health status of long- mia, tumor, or similar illness diag- lapse, and preparatory regimens for
term childhood cancer survivors has not nosed during childhood or adoles- bone marrow transplantation (if appli-
been widely studied. 13-15 Cancer- cence. This report from the CCSS is cable). Qualitative information was ab-
related medical sequelae represent only restricted to those adults (ⱖ18 years of stracted from the medical record for 42
1 of many factors contributing to sur- age) who participated in the study and specific chemotherapeutic agents; quan-
vivors’ health status. The childhood met the following eligibility criteria: (1) titative information was abstracted from
cancer experience also may produce diagnosis of leukemia, central nervous 22 of these agents. Data also were ob-
chronic psychological and cognitive im- system (CNS) tumors (all histologies), tained on surgical procedures per-
pairments that hinder posttreatment ad- Hodgkin disease, non-Hodgkin lym- formed from the time of diagnosis on-
justment and adversely affect quality of phoma, kidney tumor, neuroblastoma, ward, on tumor site, and fields and
life. Although the majority of studies soft tissue sarcoma, or bone tumor; (2) doses of radiation therapy. Copies of the
indicate that psychosocial function- diagnosis and initial treatment at 1 of 25 baseline questionnaire and the treat-
ing in childhood cancer survivors is collaborating CCSS institutions; (3) diag- ment abstraction form are available for
very good, 10% to 20% of individuals nosis date between January 1, 1970, and review and downloading at http://
show signs of psychological maladjust- December 31, 1986; (4) age younger www.cancer.umn.edu/ccss.
ment, manifested as mood distur- than 21 years at diagnosis; and (5) sur-
bances, behavioral problems, and so- vival at least 5 years from diagnosis. The Health Status Measures
matic distress. 16-19 Higher levels of CCSS protocol and contact documents Six domains of health status were mea-
psychological distress in childhood can- were reviewed and approved by the sured for this study: general health, men-
cer patients have been associated with human subjects committee at each par- tal health, functional status, limita-
academic underachievement, under- ticipating institution and informed con- tions of activity, pain as a result of the
employment, and functional limita- sent was obtained for study participa- cancer or its treatment, and anxiety/
tions that may adversely affect health tion. A detailed description of the fears as a result of the cancer or its treat-
status.16-19 General anxiety and fears methods and cohort characteristics have ment. For general health, participants
may persist after cancer treatment and been reported previously.22 were asked, “Would you say that your
undergo periodic exacerbations.20,21 In Of the 20304 childhood cancer sur- health is excellent, very good, good, fair,
extreme situations, intrusive cancer- vivors included in the cohort, 2996 or poor?” The 18-item Brief Symptom
related memories may trigger intense (14.8%) could not be located and were Inventory (BSI-18), a self-report mea-
emotional and physiological reactions considered lost to follow-up. Among the sure of psychological symptoms, was
akin to posttraumatic stress reac- 17 308 participants located, 14 193 used for the mental health domain.23,24
tions.20,21 Therefore, health status as- (82%) completed a baseline question- Responses were scored according to the
sessment should consider general naire, including 9535 of participants published manual with each partici-
health, mental health, and functional who were alive and 18 years or older pant receiving a T-score (mean = 50,
limitations resulting from both medi- at time of interview. SD=10) on the Global Severity Index,
cal and psychosocial sequelae. To allow comparisons with a repre- as well as the 3-symptom specific sub-
The goal of this study was to de- sentative noncancer population, a ran- scales—depression, somatization, and
scribe the health status of long-term sur- dom sample of participating survivors anxiety.23,24 For each scale a T-score of
vivors of childhood cancer. Using a were chosen and asked to identify their 63 or higher represents the upper 10th
large, retrospective cohort of young nearest age living sibling. From those percentile of scores reported in a com-
adult survivors of childhood cancer liv- identified, 3585 siblings to date have munity sample and is considered to be
ing at the time of interview, the 3 ob- been interviewed, of whom 2916 were significantly elevated. Using this cut-
jectives of this cross-sectional study aged 18 years or older at interview and off, paticipants’ scores were classified as
were to (1) determine the prevalence included in this analysis. Demograph- elevated vs not elevated on each of the
of adverse health status, (2) compare ics of the survivors and siblings are pro- 4 scales. Participants who had a signifi-
the health status of survivors with that vided in TABLE 1. cant elevation on any of the 3 symp-
of a cohort of siblings, and (3) identify tom specific subscales were classified as
factors associated with adverse health Cancer Treatment Information having adverse mental health, and this
status in survivors. Information on the characteristics of the was used as the primary mental health
original cancer diagnosis was ob- outcome for this analysis.
METHODS tained on all eligible cases from the Questions assessing general health,
Patient Selection and Contact treating institution. For all CCSS par- functional status, and limitations of ac-
The Childhood Cancer Survivor Study ticipants who returned a signed medi- tivity were adapted from the National
(CCSS) included individuals from many cal release, information on primary can- Health Interview Survey and the Be-
institutions who survived for 5 or more cer therapy was collected, including havioral Risk Factor Surveillance Sys-
1584 JAMA, September 24, 2003—Vol 290, No. 12 (Reprinted) ©2003 American Medical Association. All rights reserved.

Downloaded From: http://jama.jamanetwork.com/ by a Nanyang Technological University User on 05/19/2015

 HEALTH STATUS OF ADULT LONG-TERM CHILDHOOD CANCER SURVIVORS

tem Survey Questionnaire.25,26 Func- relatively few black non-Hispanic, and cancer diagnosis to baseline interview.
tional status was determined from 3 Hispanic survivors, they were com- Cancer treatment variables included in
questions that asked respondents if they bined with other race/ethnic minori- the analysis were surgery (yes/no), ra-
had any impairment or health prob- ties and analyzed as a single group. Can- diation therapy (yes/no; radiation field
lem that resulted in (1) needing “help cer-related variables included cancer including brain/head, chest/mantle, ab-
with personal care needs, such as eat- type, age at diagnosis, and interval from domen, pelvis), and chemotherapy
ing, bathing, dressing, or getting around
your home”; (2) needing “help in han- Table 1. Demographics of Adult Survivors of Childhood Cancer and Siblings*
dling routine needs, such as everyday Variables Survivors, No. (%) Siblings, No. (%)
household chores, doing necessary Sex n = 9535 n = 2916
business, shopping, or getting around Female 4452 (46.7) 1537 (52.7)
for other purposes”; or (3) “keeping you Male 5083 (53.3) 1379 (47.3)
from holding a job or attending school.” Race/ethnicity n = 9511 n = 2814
Activity status was determined from 3 White, non-Hispanic 8312 (87.4) 2589 (92.0)
questions that asked respondents if in Black, non-Hispanic 501 (5.3) 91 (3.2)
the last 2 years their health was lim- Hispanic 425 (4.5) 68 (2.4)
ited for more than 3 months in (1) the Other 273 (2.9) 66 (2.4)
kinds or amounts of moderate activi- Education† n = 9017 n = 2790
ties you can do, like moving a table, car- High school or less 2998 (33.3) 685 (24.6)
rying groceries, or bowling; (2) walk- High school + some college 6019 (66.8) 2105 (75.5)
ing upstairs or climbing a few flights of Household income n = 8362 n = 2658
stairs; or (3) walking 1 block. For pain ⬍$20 000 1893 (22.6) 336 (12.6)
as a result of the cancer or its treat- ⱖ$20 000 6469 (77.4) 2322 (87.4)
ment, survivors were asked, “Do you Health insurance n = 9374 n = 2884
currently have pain as a result of your No 1448 (15.5) 311 (10.8)
cancer or its treatment?” with re- Yes or Canadian 7926 (84.6) 2573 (89.2)
sponses recoded as a dichotomous vari- Cancer diagnosis n = 9535 NA
able. Participants reporting no pain or Leukemia 2865 (30.1) NA
a small amount of pain were com- Central nervous system 1186 (12.4) NA
pared with those with a medium Hodgkin disease 1666 (17.5) NA
amount of pain, a lot of pain, or very Non-Hodgkin lymphoma 867 (9.1) NA
bad excruciating pain. In a similar fash- Wilms tumor 636 (6.7) NA
ion, survivors were asked about anxiety/ Neuroblastoma 403 (4.2) NA
fears as a result of the cancer or its treat- Sarcoma 902 (9.5) NA
ment and responses were recorded as Bone 1010 (10.6) NA
a dichotomous variable of no or small Cancer treatment n = 8214 NA
amount of anxiety/fears vs medium Surgery only 604 (7.3) NA
amount, a lot of, or very many, ex- Radiation only 23 (0.3) NA
Chemotherapy only 340 (4.1) NA
treme anxiety/fears. Each pair of the 6
Chemotherapy + radiation 959 (11.6) NA
outcomes was correlated, with weak
Chemotherapy + surgery 1362 (16.5) NA
correlations ranging from 0.12 to 0.41.
Radiation + surgery 1172 (14.2) NA
Siblings were asked the same ques-
Chemotherapy + radiation + surgery 3754 (45.5) NA
tions as survivors regarding general
Age at interview, y
health, mental health, activities, and Mean (SD) 26.8 (6.2) 29.2 (7.4)
functional impairment, but they were Range 18-48 18-56
not asked the cancer-related ques- Age at cancer diagnosis, y NA
tions related to pain or anxiety/fears. Mean (SD) 10.0 (5.6) NA
Range 0.1-20.9 NA
Independent Variables Interval from diagnosis, y NA
Demographic and socioeconomic vari- Mean (SD) 17.4 (4.6) NA
ables considered in the analysis in- Range 6-29 NA
cluded age at time of interview, sex, race Abbreviation: NA, not applicable.
*No. (%) is based on the total participants with available data for each variable. Percentages may not equal 100 due to
and ethnicity, highest level of educa- rounding.
tional attainment, health insurance, and †High school or less means some high school or high school graduate; high school + some college means high school
graduate with either some college courses or other training.
household income. Because there were
©2003 American Medical Association. All rights reserved. (Reprinted) JAMA, September 24, 2003—Vol 290, No. 12 1585

Downloaded From: http://jama.jamanetwork.com/ by a Nanyang Technological University User on 05/19/2015

 HEALTH STATUS OF ADULT LONG-TERM CHILDHOOD CANCER SURVIVORS

(yes/no and cumulative dose of anthra- health status domain serving as the de- Univariate analyses were performed
cyclines, alkylating agents, bleomycin, pendent variable were used to compare to describe associations among demo-
and epipodophyllotoxins). the odds of adverse outcomes between graphic and cancer-related variables
survivors and siblings, adjusting for age, with the health status measures. To de-
Analysis sex, and race. Generalized estimating termine the strength of association be-
The prevalence of adverse outcomes in equations for correlated data were used tween the outcome variables and the
each of the health status domains was de- to account for potential within-family demographic and cancer-related fac-
termined for survivors and siblings. Mul- correlation between the survivor and his/ tors that were hypothesized to be im-
tiple logistic regression models with each her sibling from the same family.27 portant a priori, multiple logistic re-

Table 2. Siblings and Adult Survivors of Childhood Cancer Who Reported Moderate to Extreme Adverse Health Status*
Siblings, % Survivors, %

General Mental Functional Activity General Mental Functional Activity Any

Variables No. Health Health† Impairment Limitations No. Health Health† Impairment Limitations Pain‡ Anxiety§ Domain
Total population 2916 4.9 10.2 2.5 5.6 9535 10.9 17.2 12.0 12.5 10.2 13.2 43.6
Age at interview, y
18-24 899 4.8 11.9 2.4 4.5 3917 9.5 17.8 11.9 10.6 8.5 12.5 45.1
25-29 673 3.6 10.0 2.2 5.8 2541 10.9 17.0 11.9 13.0 9.7 13.3 40.8
30-34 625 4.7 10.8 1.9 5.6 1871 12.1 17.0 12.4 13.5 12.8 13.4 43.6
ⱖ35 719 6.3 8.0 3.4 6.7 1206 13.7 16.5 12.0 16.4 12.8 15.4 45.3
Sex
Male 1378 3.9 9.6 2.2 4.2 5083 10.2 16.2 10.6 9.9 9.8 10.6 39.7
Female 1537 5.8 10.6 2.8 6.8 4452 11.7 18.4 13.7 15.5 10.7 16.2 48.0
Race/ethnicity
White, non-Hispanic 2589 8.2 12.8 4.5 6.7 8312 10.4 17.2 11.5 12.4 10.0 13.3 43.2
Minorities㛳 225 4.6 9.9 2.4 5.4 1199 14.5 17.2 15.1 14.0 11.7 12.5 46.5
Education¶
High school or less 685 7.6 12.4 3.7 7.3 2998 16.1 20.6 21.5 16.5 12.1 13.9 55.5
High school + 2105 3.8 9.1 1.8 4.9 6019 8.0 15.5 6.2 10.0 9.1 13.0 37.5
some college
Household income
⬍$20 000 336 8.1 17.3 7.2 7.8 1893 19.5 26.1 22.8 19.6 15.2 17.1 57.6
ⱖ$20000 2322 4.4 9.3 1.5 5.1 6469 7.8 15.3 8.0 10.1 8.7 12.5 39.2
Health insurance
No 311 8.4 15.9 3.6 5.8 1448 14.7 23.3 12.8 13.0 12.9 14.5 48.8
Yes or Canadian 2573 4.5 9.4 2.4 5.6 7926 10.0 16.0 11.7 12.5 9.7 13.0 42.5
*Questions about pain or anxiety/fears as a result of the cancer were not applicable to the siblings. Because these 2 domains were not included for siblings, the “Any Domain” was
not applicable. No. (%) is based on the total with available data for each variable.
†Mental health is adverse outcome (T-score ⱖ63) in any of the 3 Brief Symptom Inventory 18-item subscales (depression, somatization, or anxiety).
‡Pain as a result of the cancer or its treatment.
§Anxiety/fears as a result of the cancer or its treatment.
㛳Minorities included black, non-Hispanic, and Hispanic and other.
¶High school or less means some high school or high school graduate; high school + some college means high school graduate with either some college courses or other training.

Table 3. Moderate to Extreme Adverse Health Status Outcomes in Adult Survivors of Childhood Cancer Compared With Siblings, Adjusted for
Age, Sex, and Race
Odds Ratio (95% Confidence Interval) BSI-18 Scales, Odds Ratio (95% Confidence Interval)

General Functional Activity Mental Depression Somatization Anxiety

Cancer Diagnosis Health Impairment Limitations Health* Global Domain Domain Domain
Total population 2.5 (2.1-3.0) 5.2 (4.1-6.6) 2.7 (2.3-3.3) 1.8 (1.6-2.1) 2.2 (1.8-2.8) 1.7 (1.4-2.0) 2.2 (1.8-2.7) 1.9 (1.5-2.4)
Leukemia 2.2 (1.8-2.8) 3.8 (2.9-5.0) 1.8 (1.5-2.3) 1.7 (1.5-2.1) 2.1 (1.6-2.7) 1.6 (1.3-2.0) 1.9 (1.5-2.5) 2.0 (1.5-2.6)
Central nervous system 3.5 (2.7-4.4) 18.0 (13.8-23.6) 4.1 (3.3-5.1) 2.0 (1.6-2.4) 2.5 (1.9-3.4) 2.2 (1.7-2.8) 2.1 (1.6-2.9) 1.7 (1.3-2.4)
Hodgkin disease 2.7 (2.2-3.4) 2.4 (1.8-3.3) 2.1 (1.7-2.7) 2.0 (1.6-2.4) 2.3 (1.8-3.0) 1.7 (1.3-2.1) 2.5 (1.9-3.2) 1.9 (1.4-2.5)
Non-Hodgkin lymphoma 2.3 (1.7-3.2) 3.0 (2.1-4.4) 2.0 (1.5-2.6) 1.8 (1.4-2.3) 2.2 (1.6-3.1) 1.8 (1.4-2.4) 1.9 (1.3-2.6) 1.8 (1.3-2.6)
Wilms tumor 1.8 (1.3-2.7) 3.2 (2.1-4.8) 2.0 (1.4-2.8) 1.3 (1.0-1.7) 1.7 (1.1-2.6) 1.1 (0.7-1.6) 1.7 (1.1-2.5) 1.7 (1.1-2.6)
Neuroblastoma 2.1 (1.3-3.2) 3.8 (2.3-6.2) 2.7 (1.9-4.0) 1.4 (1.0-2.0) 1.3 (0.8-2.1) 1.4 (0.9-2.1) 1.6 (1.0-2.6) 1.5 (0.9-2.4)
Sarcoma 2.2 (1.6-2.9) 4.2 (3.1-5.9) 2.3 (1.8-3.0) 1.7 (1.3-2.1) 2.1 (1.5-2.9) 1.5 (1.1-2.0) 2.1 (1.6-2.9) 1.7 (1.2-2.4)
Bone 2.6 (2.0-3.4) 7.4 (5.5-10.0) 6.4 (5.2-8.0) 1.8 (1.5-2.3) 2.2 (1.6-3.0) 1.6 (1.2-2.1) 2.2 (1.6-3.0) 2.1 (1.5-2.8)
Abbreviation: BSI-18, Brief Symptom Inventory 18-item.
*Mental health is adverse outcome (T-score ⱖ63) in any of the 3 BSI-18 subscales.

1586 JAMA, September 24, 2003—Vol 290, No. 12 (Reprinted) ©2003 American Medical Association. All rights reserved.

Downloaded From: http://jama.jamanetwork.com/ by a Nanyang Technological University User on 05/19/2015

 HEALTH STATUS OF ADULT LONG-TERM CHILDHOOD CANCER SURVIVORS

gression models were constructed RESULTS more, and some form of health insur-
estimating odds ratios (ORs) with 95% The mean age at interview for survi- ance (Table 1).
confidence intervals (CIs) for adverse vors was 26.8 years (range, 18-48 years). TABLE 2 shows the percentage of sur-
health status. Compared with survi- The mean age at cancer diagnosis was vivors and siblings with an adverse out-
vors with a higher prevalence of late ef- 10.0 years (range, 0.1- 20.9 years), with come in the different health status do-
fects (CNS tumor, bone tumor) and a mean and median interval from diag- mains for various demographic
those with a lower prevalence (Wilms nosis to completion of questionnaire of categories. Overall, 43.6% of survivors
tumor, neuroblastoma), leukemia sur- 17.4 years and 17.2 years (range, 6-29 had at least 1 adversely affected do-
vivors are generally considered an in- years). Forty-seven percent of the par- main. TABLE 3 shows the OR estimates
termediate risk group. Thus, for regres- ticipating survivors were female, and with 95% CIs for the 4 comparative do-
sion analyses related to type of cancer, 87.4% were white, non-Hispanic. Com- mains, adjusted for age, sex, and race.
leukemia survivors were used as the ref- pared with the siblings, survivors were The percentage of survivors with ad-
erence group. Statistical significance younger, more likely to be male, and verse health status, by domain, for dif-
was set as P⬍.05. Data were analyzed less likely to be white, non-Hispanic, ferent cancer and cancer treatment vari-
with the SAS PC software package ver- to have a higher level of education, a ables are provided in TABLE 4. Multiple
sion 8 (SAS Institute, Cary, NC). household income level of $20000 or regression model fits including demo-

Table 4. Percentage of Adult Survivors of Childhood Cancer With Adverse Health Status, by Cancer Diagnosis and Cancer Treatment*
General Mental Functional Activity Any
Variables No. Health Health† Impairment Limitations Pain‡ Anxiety§ Domain
Total population 9535 10.9 17.2 12.0 12.5 10.2 13.2 43.6
Cancer diagnosis
Leukemia 2865 9.6 17.5 9.3 8.6 8.3 11.9 40.3
Central nervous system 1186 14.6 19.0 31.7 17.8 9.3 12.1 54.2
Hodgkin disease 1666 12.7 17.8 6.4 11.3 6.6 15.9 40.2
Non-Hodgkin lymphoma 867 9.6 17.2 7.2 9.0 7.3 11.6 37.4
Wilms tumor 636 8.2 14.1 7.7 8.8 8.4 11.2 37.5
Neuroblastoma 403 8.6 15.6 8.3 11.7 7.6 10.7 41.2
Sarcoma 902 9.9 16.1 9.8 11.3 14.8 16.0 43.2
Bone 1010 12.1 17.0 16.2 26.8 23.0 15.2 55.7
Age at diagnosis, y
0-4 2409 9.4 16.9 12.9 10.7 8.5 11.7 44.1
5-9 2359 10.1 17.7 12.4 10.6 8.6 12.7 42.6
10-14 2577 11.4 17.7 11.3 13.4 11.1 13.9 43.4
15-21 2190 12.8 16.5 11.6 15.7 12.9 14.7 44.4
Radiotherapy
Any 5925 11.4 17.8 12.6 11.9 9.5 13.9 44.0
Brain 2632 11.3 17.9 17.6 11.4 8.8 12.3 46.5
Chest 1912 11.8 18.1 7.4 12.3 8.9 15.6 42.2
Abdomen 1076 10.2 17.4 8.2 11.7 9.4 13.5 40.5
Pelvis 787 11.6 19.1 10.7 12.4 11.7 13.9 42.2
Chemotherapy
Any chemotherapy 6434 10.5 17.8 10.7 11.8 9.8 13.8 43.4
Alkylating agent
None 2181 8.3 16.3 9.8 9.7 8.3 11.3 40.6
Intensity
Low 1590 11.3 18.9 10.5 12.8 11.7 13.9 45.1
Moderate 1173 10.5 17.5 10.4 12.1 11.4 14.4 43.6
High 886 12.0 18.7 11.2 11.4 7.7 16.1 41.1
Anthracycline
None 3320 10.1 17.5 10.8 9.7 7.4 12.9 41.5
Intensity
Low 831 11.9 18.5 9.9 10.4 11.0 13.5 42.9
Moderate 970 10.5 19.1 10.3 15.0 11.5 14.4 46.1
High 1058 10.3 16.7 10.6 14.9 14.1 16.5 45.0
*No. (%) is based on the total participants with available data for each variable.
†Mental health is adverse outcome (T-score ⱖ63) in any of the 3 Brief Symptom Inventory 18-item subscales (depression, somatization, or anxiety).
‡Pain as a result of the cancer or its treatment.
§Anxiety/fears as a result of the cancer or its treatment.

©2003 American Medical Association. All rights reserved. (Reprinted) JAMA, September 24, 2003—Vol 290, No. 12 1587

Downloaded From: http://jama.jamanetwork.com/ by a Nanyang Technological University User on 05/19/2015

 HEALTH STATUS OF ADULT LONG-TERM CHILDHOOD CANCER SURVIVORS

graphic and cancer-related factors as- and sarcomas (OR, 1.2; 95% CI, 1.1- CI, 3.3-4.5; P⬍.001) and activity sta-
sociated with adverse outcomes in each 1.5; P= .01). tus (OR, 1.9; 95% CI, 1.6-2.2; P⬍.001).
of the domains are provided in TABLE 5. Being female was associated with ad- Similarly, an annual household in-
Reporting at least 1 adversely affected verse outcomes in all of the domains ex- come of less than $20000 was associ-
health status domain was associated cept for cancer-related pain. Survivors ated with adverse health status in all do-
with being female (OR, 1.4; 95% CI, aged 35 years or older were more likely mains, especially functional status (OR,
1.3-1.6; P⬍.001), not completing high to report adverse outcomes in general 2.9; 95% CI, 2.5-3.4; P⬍.001) and gen-
school or being a high school gradu- health (OR, 1.7; 95% CI, 1.4-2.2; eral health (OR, 2.6; 95% CI, 2.2-3.0;
ate without further training or educa- P⬍.001), functional status (OR, 1.4; P⬍.001). Of note, racial and ethnic mi-
tion (OR, 2.0; 95% CI, 1.8-2.2; P⬍.001), 95% CI, 1.1-1.8; P=.03), activity sta- nority status as a group was associated
and an annual household income of less tus (OR, 1.6; 95% CI, 1.3-1.9; P⬍.001), with adverse outcomes in general health
than $20 000 (OR, 1.8; 95% CI, 1.6- and pain as a result of the cancer or its (OR, 1.2; 95% CI, 1.1-1.5; P =.04), but
2.1; P⬍.001). Compared with leuke- treatment (OR, 1.6; 95% CI, 1.3-2.1; not in any of the other domains. There
mia survivors, increased risks for an P⬍.001), compared with survivors aged were 2266 survivor-sibling pairs in the
adverse health status was noted in 18 to 24 years. A lower level of educa- study. When analyzing only these pairs,
survivors of bone tumors (OR, 2.1; 95% tional attainment was associated with excluding the 76% of survivors with-
CI, 1.8-2.5; P⬍.001), CNS tumors adverse outcomes in each domain, es- out a sibling, the conclusions were un-
(OR, 1.7; 95% CI, 1.5-2.0; P⬍.001), pecially functional status (OR, 3.9; 95% changed (data not shown).

Table 5. Multiple Regression Results of Demographic and Socioeconomic Factors for Adverse Health Status, by Domain in 9535 Adult
Survivors of Childhood Cancer
Odds Ratio (95% Confidence Interval)

General Mental Functional Activity Any

Variables Health Health* Impairment Limitations Pain† Anxiety‡ Domain
Age at interview, y
18-24 1.0 1.0 1.0 1.0 1.0 1.0 1.0
25-29 1.3 (1.1-1.5) 1.0 (0.8-1.1) 1.2 (1.0-1.4) 1.3 (1.1-1.5) 1.2 (1.0-1.4) 1.1 (0.9-1.2) 0.9 (0.8-1.0)
30-34 1.5 (1.2-1.8) 1.0 (0.8-1.1) 1.2 (1.0-1.5) 1.2 (1.0-1.5) 1.6 (1.3-1.9) 1.0 (0.8-1.2) 1.0 (0.8-1.1)
ⱖ35 1.7 (1.4-2.2) 0.9 (0.8-1.1) 1.4 (1.1-1.8) 1.6 (1.3-1.9) 1.6 (1.3-2.1) 1.1 (0.9-1.3) 1.0 (0.9-1.2)
Sex
Male 1.0 1.0 1.0 1.0 1.0 1.0 1.0
Female 1.2 (1.0-1.4) 1.2 (1.1-1.3) 1.4 (1.2-1.6) 1.7 (1.5-1.9) 1.1 (0.9-1.2) 1.6 (1.4-1.9) 1.4 (1.3-1.6)
Race/ethnicity
White, non-Hispanic 1.0 1.0 1.0 1.0 1.0 1.0 1.0
Minorities§ 1.2 (1.1-1.5) 0.9 (0.8-1.1) 1.1 (0.9-1.4) 1.0 (0.8-1.2) 1.0 (0.9-1.3) 0.9 (0.7-1.1) 1.0 (0.9-1.1)
Education㛳
High school or less 2.0 (1.7-2.3) 1.3 (1.1-1.5) 3.9 (3.3-4.5) 1.9 (1.6-2.2) 1.4 (1.2-1.6) 1.1 (1.0-1.3) 2.0 (1.8-2.2)
High school + some college 1.0 1.0 1.0 1.0 1.0 1.0 1.0
Household income
⬍$20 000 2.6 (2.2-3.0) 1.8 (1.5-2.0) 2.9 (2.5-3.4) 2.0 (1.8-2.4) 1.8 (1.5-2.1) 1.4 (1.2-1.7) 1.8 (1.6-2.1)
ⱖ$20 000 1.0 1.0 1.0 1.0 1.0 1.0 1.0
Health insurance
No 1.2 (1.0-1.4) 1.4 (1.2-1.6) 0.8 (0.6-0.9) 0.9 (0.7-1.1) 1.2 (1.0-1.5) 1.1 (0.9-1.3) 1.1 (1.0-1.2)㛳
Yes or Canadian 1.0 1.0 1.0 1.0 1.0 1.0 1.0
Cancer diagnosis
Leukemia 1.0 1.0 1.0 1.0 1.0 1.0 1.0
Central nervous system 1.5 (1.2-1.8) 1.1 (0.9-1.3) 4.7 (3.9-5.7) 2.2 (1.8-2.7) 1.0 (0.8-1.3) 1.0 (0.8-1.3) 1.7 (1.5-2.0)
Hodgkin disease 1.4 (1.2-1.8) 1.1 (1.0-1.4) 0.8 (0.6-1.0) 1.4 (1.1-1.8) 0.7 (0.6-0.9) 1.4 (1.2-1.8) 1.1 (1.0-1.3)㛳
Non-Hodgkin lymphoma 1.0 (0.8-1.4) 1.0 (0.8-1.3) 0.9 (0.6-1.2) 1.2 (0.9-1.6) 0.8 (0.6-1.1) 1.1 (0.8-1.4) 1.0 (0.8-1.2)
Wilms tumor 0.9 (0.6-1.2) 0.8 (0.6-1.0) 0.8 (0.6-1.1) 1.0 (0.8-1.4) 1.1 (0.8-1.5) 0.9 (0.7-1.2) 0.9 (0.7-1.1)
Neuroblastoma 1.0 (0.7-1.4) 0.9 (0.6-1.2) 1.0 (0.6-1.4) 1.5 (1.1-1.8) 1.0 (0.7-1.5) 0.9 (0.6-1.2) 1.1 (0.8-1.3)
Sarcoma 1.1 (0.8-1.4) 1.0 (0.8-1.2) 1.2 (0.9-1.6) 1.4 (1.1-1.8) 1.9 (1.5-2.3) 1.4 (1.2-1.8) 1.2 (1.1-1.5)§
Bone 1.3 (1.0-1.7) 1.0 (0.9-1.3) 2.3 (1.8-2.9) 4.1 (3.4-5.1) 3.1 (2.5-3.8) 1.4 (1.1-1.7) 2.1 (1.8-2.5)
*Mental health is adverse outcome (T-score ⱖ63) in any of the 3 Brief Symptom Inventory 18-item subscales (depression, somatization, or anxiety).
†Pain as a result of the cancer or its treatment.
‡Anxiety/fears as a result of the cancer or its treatment.
§Minorities included black non-Hispanic, Hispanic, and other.
㛳High school or less means some high school or high school graduate; high school + some college means high school graduate with either some college courses or other training.

1588 JAMA, September 24, 2003—Vol 290, No. 12 (Reprinted) ©2003 American Medical Association. All rights reserved.

Downloaded From: http://jama.jamanetwork.com/ by a Nanyang Technological University User on 05/19/2015

 HEALTH STATUS OF ADULT LONG-TERM CHILDHOOD CANCER SURVIVORS

Compared with leukemia survivors, affected health status compared with sur- of adults surviving childhood cancer.
adjusting for demographic/socioeco- vivors of leukemia (Table 5). No inter- Through the resource of the CCSS, a
nomic variables, survivors of bone tu- action was observed between chemo- more precise and comprehensive as-
mor were more likely to report adverse therapy and radiation therapy with any sessment of health status was possible
health in each domain (except mental of the outcomes. Of note, era of diag- because of the number and diversity of
health), especially activity status (OR, nosis was not significantly associated study participants and the varied ex-
4.1; 95% CI, 3.4-5.1; P⬍.001), func- with any of the adverse health status pertise of the investigators. Important
tional status (OR, 2.3; 95% CI, 1.8-2.9; outcomes other than functional im- study findings include the general
P⬍.001), and pain as a result of the can- pairment. health as perceived by adults surviv-
cer or its treatment (OR, 3.1; 95% CI, Within the survivor cohort, 313 had ing childhood cancer is very good with
2.5-3.8; P⬍.001). Similarly, survivors of a confirmed second malignant primary only 10.9% reporting fair or poor
CNS tumor were more likely to report neoplasm, excluding nonmelanoma skin health, long-term adverse effects in spe-
adverse outcomes in functional status cancer. Compared with survivors who cific aspects of health were relatively
(OR, 4.7; 95% CI, 3.9-5.7; P⬍.001), ac- did not have a second malignant pri- common as reflected by 43.6% of the
tivity status (OR, 2.2; 95% CI, 1.8-2.7; mary neoplasm, this subset of survi- cohort reporting impairment in 1 or
P⬍.001), and general health (OR, 1.5; vors were more likely to report moder- more of the health domains evaluated
95% CI, 1.2-1.8; P=.002). ate to extreme adverse outcomes for each in the study, and factors associated with
The relationship of treatment to health health status domain: any domain impaired health status included being
domains is summarized in TABLE 6. (OR,1.8; 95% CI, 1.4-2.3; P⬍.001); gen- female, not completing high school,
Treatment with surgery (OR, 1.2; 95% eral health (OR, 1.8; 95% CI, 1.3-2.6; having a household income less than
CI, 1.1-1.4; P=.01), radiation involv- P⬍.001); mental health (OR, 1.7; 95% $20000, and having a diagnosis of bone
ing the head/brain (OR, 1.4; 95% CI, 1.2- CI, 1.3-2.3; P⬍.001); functional impair- tumor, CNS tumor, sarcoma, or Hodg-
1.6; P⬍.001), chest/mantle radiation ment (OR, 2.0; 95% CI, 1.3-3.0; kin disease. These findings help char-
(OR, 1.2; 95% CI, 1.1-1.4; P=.05), or P⬍.001); activity limitations (OR, 1.8; acterize the high-risk childhood can-
alkylating agent chemotherapy (OR, 1.2; 95% CI, 1.3-2.6; P⬍.001); pain as a result cer survivor who is more likely to
95% CI, 1.1-1.4; P=.02) was associated of cancer (OR. 1.5; 95% CI, 1.1-2.3; require intervention to optimize long-
with a mild excess risk of reporting at P=.03); and anxiety as a result of the can- term health outcomes.
least 1 adversely affected health status cer (OR, 1.6; 95% CI, 1.2-2.2; P=.002). Previous reports of global health
domain. Even after accounting for treat- have less relevance to the general
ment effects, survivors of both CNS and COMMENT population of childhood cancer survi-
bone tumors had an approximately This study provides information about vors because most describe outcomes
2-fold odds of having at least 1 adversely the health status of the largest cohort in small, heterogeneous patient popu-

Table 6. Multiple Regression Results of Cancer-Related Risk Factors for Adverse Health Status, by Domain, Adjusted for Age, Sex, Race, and
Diagnosis
Odds Ratio (95% Confidence Interval)

General Mental Functional Activity Any

Variable Health Health* Impairment Limitations Pain† Anxiety‡ Domain
Surgery
No 1.0 1.0 1.0 1.0 1.0 1.0 1.0
Yes 1.2 (0.9-1.5) 1.1 (0.9-1.3) 1.2 (0.9-1.5) 1.1 (0.9-1.4) 1.3 (1.0-1.7) 1.2 (1.0-1.6) 1.2 (1.1-1.4)
Radiation therapy
None 1.0 1.0 1.0 1.0 1.0 1.0 1.0
Head/brain 1.5 (1.2-1.8) 1.0 (0.8-1.2) 2.1 (1.7-2.6) 1.3 (1.0-1.6) 1.5 (1.1-1.9) 1.1 (0.9-1.3) 1.4 (1.2-1.6)
Chest/mantle 1.4 (1.0-1.8) 1.1 (0.9-1.3) 1.1 (0.8-1.4) 1.3 (1.0-1.6) 1.4 (1.1-1.9) 1.2 (0.9-1.5) 1.2 (1.1-1.4)
Brain/chest 1.4 (0.8-2.5) 1.1 (0.7-1.7) 3.6 (2.2-5.9) 2.0 (1.2-3.4) 3.4 (2.1-5.5) 1.2 (0.7-2.0) 2.1 (1.4-3.0)
Other 1.4 (1.1-1.8) 1.1 (0.9-1.3) 1.4 (1.1-1.8) 1.2 (1.0-1.5) 1.5 (1.2-1.9) 1.2 (1.0-1.5) 1.2 (1.0-1.4)
Chemotherapy
None 1.0 1.0 1.0 1.0 1.0 1.0 1.0
Alkylating agent 1.3 (1.1-1.7) 1.2 (1.0-1.4) 1.4 (1.1-1.8) 1.1 (0.9-1.4) 1.0 (0.8-1.3) 1.2 (1.0-1.5) 1.2 (1.1-1.4)
Anthracycline 1.2 (0.9-1.7) 1.1 (0.8-1.5) 1.5 (1.1-2.1) 1.3 (1.0-1.8) 1.2 (0.9-1.7) 1.1 (0.8-1.5) 1.2 (0.9-1.5)
Alkylating agent + anthracycline 1.4 (1.1-1.8) 1.2 (1.0-1.5) 1.5 (1.1-1.9) 1.4 (1.1-1.8) 1.3 (1.0-1.7) 1.5 (1.2-1.9) 1.3 (1.1-1.5)
Other 1.1 (0.8-1.4) 1.0 (0.8-1.3) 1.4 (1.1-2.0) 1.1 (0.8-1.5) 0.9 (0.6-1.2) 1.1 (0.9-1.5) 1.2 (1.0-1.4)
*Mental health is adverse outcome (T-score ⱖ63) in any of the 3 Brief Symptom Inventory 18-item subscales (depression, somatization, or anxiety).
†Pain as a result of the cancer or its treatment.
‡Anxiety/fears as a result of the cancer or its treatment.

©2003 American Medical Association. All rights reserved. (Reprinted) JAMA, September 24, 2003—Vol 290, No. 12 1589

Downloaded From: http://jama.jamanetwork.com/ by a Nanyang Technological University User on 05/19/2015

 HEALTH STATUS OF ADULT LONG-TERM CHILDHOOD CANCER SURVIVORS

lations, who did not receive contem- sequelae on health status of long-term tumors who experienced a variety of
porary antineoplastic treatment.13-15 childhood cancer survivors. disabling neurocognitive complica-
Garre et al14 evaluated the health sta- Sociodemographic factors that were tions related to tumor location and CNS
tus of 288 survivors treated for child- consistently associated with adverse therapy. Among all disease groups stud-
hood cancer from 1962 to 1982 at a outcomes across all health domains in- ied, adverse health outcomes in cancer-
single institution using medical and cluded being female, lower levels of related pain, functional impairment,
neuropsychological assessments. educational attainment, and house- and activity limitations were associ-
Health status was defined relative to hold income less than $20000. These ated with older age at interview. These
the incidence of medical sequelae and factors have been previously associ- results suggest that a higher degree of
graded according to severity and cor- ated with a higher risk of emotional se- chronic disability may develop in ag-
rective interventions, and did not quelae after childhood leukemia and ing childhood cancer survivors or that
include other aspects of health. lymphoma.16,19 Mental health prob- earlier treatment produces more mor-
Only 1 previous investigation de- lems also have been observed more fre- bidity than contemporary therapies.
scribed health status of adults surviv- quently in the general population with Moderate to severe impairment in
ing childhood cancer in the context of these same sociodemographic fea- some aspect of mental health was
medical and psychosocial sequelae and tures.28,29 Racial and ethnic minority sta- observed across all diagnostic groups ana-
health-related quality of life. Crom et tus also was associated with a slight ex- lyzed. In particular, patients with Hodg-
al13 examined the influence of psycho- cess risk of impaired general health kin disease, sarcomas, and bone tumors
social demographic factors and pa- (OR, 1.2; 95% CI 1.1-1.5), but was not had significantly higher levels of cancer-
tient and cancer-related variables on associated with adverse outcomes in any related anxiety and fears adversely affect-
health status and related quality of life of the other health domains. The use ing health status. These malignancies
in 220 adults who were 15 or more of a sibling control group permitted commonly present during adolescence
years from diagnosis of a childhood comparison of cancer survivors with a and young adulthood when patients
solid tumor. Among the group, 59.1% demographically and genetically simi- developmentally and cognitively can bet-
reported at least 1 serious toxic effect lar cohort. Predictably, survivors ex- ter appreciate the gravity of a cancer diag-
(infertility, thyroid dysfunction, sco- hibited significantly more adverse out- nosis and risks of cancer treatment
liosis most common). Despite the fre- comes in all of the domains studied sequelae. Recent studies indicate that a
quency of cancer-related toxicity, two compared with sibling controls, sug- significant minority of young adult sur-
thirds of survivors reported moder- gesting a major role of treatment ef- vivors of childhood cancer experience
ately good to excellent quality of life. fects in determining health status. symptoms of posttraumatic stress related
Health status in this group of indi- Our findings of poor health out- to residual anxiety about their cancer
viduals is likely to be determined by the comes in survivors of primary malig- experience.21 In severely affected patients,
unique response of the survivor and fam- nancies treated with more intensive external reminders of diagnosis and treat-
ily to a multitude of physical and emo- multimodality therapy concur with pre- ment may stimulate an exacerbation of
tional insults associated with the can- vious investigations of long-term child- anxiety and distress that interferes with
cer experience. While health status is hood cancer survivors.13,14 It is not sur- daily functioning. Often this anxiety is
generally poorer in survivors with more prising that survivors of pediatric bone manifest as somatic distress regarding the
frequent or serious medical sequelae, re- tumors and sarcoma treated with che- body, current state of health, and related
silient survivors with significant cancer- motherapy and aggressive surgery or ra- fears of recurrence or severe late toxic
related medical issues may paradoxi- diation more frequently reported can- effects. These issues make it difficult to
cally view their overall health as good. cer-related pain and activity limitations. determine if somatic complaints such as
Conversely, health status may be signifi- Treatment for these tumors generally fatigue, lethargy, chronic pain, or chronic
cantly impaired in survivors with less se- includes high cumulative doses of an- health worries are physical or psycho-
rious or no medical sequelae who have thracyclines and alkylating agents, logical in etiology. Therefore, all symp-
psychosocial issues interfering with post- which predispose to cardiomyopathy, toms should be evaluated from a multi-
treatment psychological adaptation. To infertility, and second malignancy, as faceted viewpoint that considers physical
integrate the impact of both medical and well as surgical interventions and/or ra- and psychological issues in the differen-
psychosocial factors on survivor health diation therapy that may limit muscu- tial diagnoses.30
outcomes, in the present study we de- loskeletal development and function Appreciation of sociodemographic
fined health status in the context of 6 do- and produce chronic pain. Functional and treatment characteristics associ-
mains that assessed general health, men- impairment was more commonly en- ated with physical morbidity and psy-
tal health, functional impairment, activity dorsed in survivors of bone tumors, chosocial maladjustment is an impor-
limitations, cancer-related pain, and anxi- who encountered chronic health is- tant first step in developing interventions
ety/fears. This definition permitted the sues following amputation or limb- to improve the health status of long-
evaluation of medical and psychosocial sparing surgeries, and survivors of CNS term survivors. Health care profession-
1590 JAMA, September 24, 2003—Vol 290, No. 12 (Reprinted) ©2003 American Medical Association. All rights reserved.

Downloaded From: http://jama.jamanetwork.com/ by a Nanyang Technological University User on 05/19/2015

 HEALTH STATUS OF ADULT LONG-TERM CHILDHOOD CANCER SURVIVORS

als are challenged with the responsibil- survivors with heterogeneous cancer di- deficits in specific health domains are
ity of educating survivors about anxiety- agnoses, ages/developmental stages at di- common. Sociodemographic factors pre-
provoking cancer-related risks using agnosis, and treatments during a 16- dicting poor health are similar to those
methods that encourage continued year period. Second, health status was, identified in the general population and
health monitoring and motivate life- in part, determined by self-report of include being female, those with low in-
style practices that promote risk reduc- medical complications that were not vali- come, and those with low educational
tion. This task is particularly difficult dated externally. Third, we did not use achievement. Survivors of CNS tu-
considering the fact that many primary a single validated instrument for assess- mors, bone tumors, and sarcomas pre-
care clinicians and young-adult cancer ment of health status, but derived this dictably reported functional impair-
survivors lack knowledge about their measure from a combination of vari- ments and/or activity limitations likely
cancer treatment, its associated health ous instruments. Although we believe related to the aggressive nature of anti-
risks, and appropriate screening mea- that the health domains assessed pro- neoplastic therapy, especially those di-
sures.31 vide a reasonable measure of health sta- rected at local tumor control. Linger-
The process can be further under- tus, we recognize that the question- ing cancer-related anxiety and fears were
mined if a survivor is experiencing un- naire did not provide an entirely global more common in long-term survivors of
resolved cancer-related anxiety or post- view of health status. Fourth, incom- Hodgkin disease, sarcomas, and bone tu-
traumatic stress symptoms. Hobbie et plete participation of the eligible adult mors possibly reflecting a greater ap-
al21 recommended that clinicians pro- cohort could result in more favorable preciation of their vulnerability to can-
mote competence of young adult can- health status if healthier survivors were cer-related health risks. Primary care
cer survivors in dealing with potential more likely to join the cohort study. clinicians should anticipate health defi-
and unknown health risks after can- Conversely, health status deficits may be cits in these clinical and sociodemo-
cer. To do so effectively, health care pro- overestimated if survivors with chronic graphic groups when evaluating adults
fessionals and pediatric cancer cen- health problems participated more fre- who are childhood cancer survivors and
ters must communicate with each other quently than healthy survivors. Fifth, be- be prepared to address physical and psy-
to obtain accurate information about cause of the small proportion of minor- chosocial sequelae adversely impact-
treatment exposures and their poten- ity study participants in the CCSS, these ing health status.
tial long-term adverse effects. Peri- results have limited generalizability to
odic evaluations of survivors should in- ethnic and minority population. Author Contributions: Study concept and design:
Hudson, Mertens, Gurney, Robison, Oeffinger.
clude a thorough psychosocial and Finally, the use of a sibling cohort Acquisition of data: Hudson, Mertens, Robison,
physical assessment for cancer- also could pose problems with inter- Oeffinger.
Analysis and interpretation of data: Hudson, Mertens,
related medical and emotional se- pretation of health status outcomes that Yasui, Hobbie, Chen, Gurney, Yeazel, Recklitis, Marina,
quelae that require further interven- should be considered. A growing body Robison, Oeffinger.
tion. During the evaluation, the of literature exists describing psycho- Drafting of the manuscript: Hudson, Hobbie, Gurney,
Robison, Oeffinger.
clinician should candidly discuss can- logical outcomes in siblings of cancer Critical revision of the manuscript for important in-
cer-related health risks, correct mis- survivors.32-37 While some investiga- tellectual content: Hudson, Mertens, Yasui, Chen,
Gurney, Yeazel, Recklitis, Marina, Robison, Oeffinger.
perceptions, and encourage the survi- tors have reported that siblings of child- Statistical expertise: Mertens, Yasui, Chen, Gurney,
vor to stay informed about this rapidly hood cancer patients experience more Recklitis, Robison.
Obtained funding: Robison, Oeffinger.
evolving area of medical research. emotional and behavioral problems Administrative, technical, or material support: Hudson,
Modifiable behavioral risk factors that compared with age-matched con- Robison, Oeffinger.
Study supervision: Hudson, Oeffinger.
may exacerbate cancer-related risks trols,33,38 others indicate no significant Topic: Hobbie.
should be emphasized in an effort to differences in emotional status.34,35 Childhood Cancer Survivor Study (CCSS) Institu-
shift perspective from potential uncon- Some studies have observed positive tions and Investigators: University of California-San
Francisco, Arthur Ablin, MD (principal investigator);
trollable aspects of illness and moti- psychological outcomes in siblings of University of Alabama, Birmingham, Roger Berkow,
vate behaviors that maintain well- survivors who were noted to be more MD (principal investigator); International Epidemiol-
ogy Institute, Rockville, Md, John Boice, ScD (former
ness.21 Clinicians should likewise be thoughtful, sensitive, compassionate, principal investigator); University of Washington, Se-
prepared to assist with referrals to other mature, and responsible because of their attle, Norman Breslow, PhD (former principal inves-
tigator); University of Texas-Southwestern Medical
health care professionals that can ame- cancer experience.34,36,37 The BSI-18 Center at Dallas, George R. Buchanan, MD (principal
liorate or facilitate adaptation to physi- scores of our sibling cohort, which in- investigator); Kevin Oeffinger, MD (CCSS Steering
cal or emotional disabilities related to corporate a comparison with a norma- Committee); Dana-Farber Cancer Institute, Boston,
Mass, Lisa Diller, MD (principal investigator); Hol-
childhood cancer. tive control group, did not reflect more combe Grier, MD (former principal investigator); Fre-
This study has some methodological mental health problems than the aver- derick Li, MD (former principal investigator); Texas Chil-
dren’s Center, Houston, Zoann Dreyer, MD (principal
limitations that should be considered age person.23,24 investigator); Children’s Hospital and Medical Cen-
when interpreting the results. First, there In summary, although the vast ma- ter, Seattle, Wash, Debra Friedman, MD, MPH (prin-
cipal investigator), Thomas Pendergrass, MD (former
are inherent difficulties in describing jority of adults surviving childhood can- principal investigator); Roswell Park Cancer Institute,
health outcomes in childhood cancer cer perceive their overall health as good, Buffalo, NY, Daniel M. Green, MD (principal investi-

©2003 American Medical Association. All rights reserved. (Reprinted) JAMA, September 24, 2003—Vol 290, No. 12 1591

Downloaded From: http://jama.jamanetwork.com/ by a Nanyang Technological University User on 05/19/2015

 HEALTH STATUS OF ADULT LONG-TERM CHILDHOOD CANCER SURVIVORS

gator); (CCSS Steering Committee); Hospital for Sick A. Kim Ritchey, MD (principal investigator), Julie Blatt, fornia-Los Angeles, Lonnie Zeltzer, MD (principal in-
Children, Toronto, Ontario, Canada, Mark Green- MD (former principal investigator); University of Min- vestigator); (CCSS Steering Committee).
berg, MB, ChB (principal investigator); St. Louis Chil- nesota, Minneapolis, Leslie L. Robison, PhD (princi- Funding/Support: This work was supported by grant
dren’s Hospital, Mo, Robert Hayashi, MD (principal pal investigator); (CCSS Steering Committee); Ann 5U24-CA-55727 from the Department of Health and
investigator); Teresa Vietti, MD (former principal in- Mertens, PhD (CCSS Steering Committee); Joseph Human Services and funding to the University of Min-
vestigator); St. Jude Children’s Research Hospital, Neglia, MD, MPH (CCSS Steering Committee); Mark nesota from the Children’s Cancer Research Fund. Dr
Memphis, Tenn, Melissa Hudson, MD (principal in- Nesbit, MD (CCSS Steering Committee); Cincinnati Hudson is supported by Cancer Center Support (CORE)
vestigator); University of Michigan, Ann Arbor, Mich, Children’s Hospital Medical CenterStella Davies, MD, grant CA 21765 from the National Cancer Institute
Raymond Hutchinson, MD (principal investigator); PhD (CCSS Steering Committee); Children’s Hospital and by the American Lebanese Syrian Associated Chari-
Stanford University School of Medicine, Stanford, Calif, Los Angeles, Calif, Kathy Ruccione, RN, MPH (prin- ties (ALSAC).
Michael P. Link, MD (principal investigator); Sarah S. cipal investigator); Memorial Sloan-Kettering Cancer Previous Presentations: Presented at the Proceed-
Donaldson, MD (CCSS Steering Committee); Chil- Center New York, Charles Sklar, MD (principal inves- ings of the 2002 Annual Meeting of the American So-
dren’s Hospital of Philadelphia, Pa, Anna Meadows, tigator); (CCSS Steering Committee); National Can- ciety of Clinical Oncology, Orlando, Fla, May 18-21,
MD (principal investigator); (CCSS Steering Commit- cer Institute, Bethesda, Md, Malcolm Smith, MD (CCSS 2002; Proceedings of the 7th International Confer-
tee); Bobbie Bayton (CCSS Steering Committee); Chil- Steering Committee); Peter Inskip, ScD (CCSS Steer- ence on the Long-Term Complications of Treatment
dren’s Hospital, Oklahoma City, Okla, John Mulvi- ing Committee); Mayo Clinic, Rochester, Minn, W. An- of Children and Adolescents for Cancer, Niagra-on-
hill, MD (CCSS Steering Committee); Children’s thony Smithson, MD (principal investigator); Gerald the-Lake, Ontario, Canada, June 28-29, 2002; Con-
Hospital, Denver, Colo, Brian Greffe (principal inves- Gilchrist, MD (former principal investigator); Univer- ference on Cancer Survivorship: Resilence Across the
tigator); Lorrie Odom, MD (former principal investi- sity of Texas M.D. Anderson Cancer Center, Louise Lifespan, Washington, DC, June 2002; and Proceed-
gator); Children’s Health Care-Minneapolis, Minn, Strong, MD (principal investigator); (CCSS Steering ings of the American Institute for Cancer Reseach and
Maura O’Leary, MD (principal investigator); Colum- Committee); Marilyn Stovall, PhD (CCSS Steering Com- World Cancer Research Fund International: Interna-
bus Children’s Hospital, Ohio, Amanda Termuhlen, MD mittee); Riley Hospital for Children, Indianapolis, Ind, tional Research Conference on Food, Nutrition and
(principal investigator); Frederick Ruymann, MD Terry A. Vik, MD (principal investigator); Robert Weet- Cancer, Washington, DC, July 17-18, 2003.
(former principal investigator); Stephen Qualman, MD man, MD (former principal investigator); Fred Hutchin- Acknowledgment: We thank John Whitton, MS, and
(CCSS Steering Committee); Children’s National Medi- son Cancer Center, Seattle, Wash, Yutaka Yasui, PhD Pauline Mitby, MPH, for their assistance with the CCSS
cal Center, Washington, DC, Gregory Reaman, MD (principal investigator); (CCSS Steering Committee); database and Sandra Mulkey, RD, and Barbara
(principal investigator); Roger Packer, MD (CCSS Steer- John Potter, MD, PhD (former principal investiga- Cruchon for their assistance with manuscript prepa-
ing Committee); Children’s Hospital of Pittsburgh, Pa, tor); (CCSS Steering Committee); University of Cali- ration.

REFERENCES
1. Green DM, Zevon MA, Reese PA, Lowrie GS, Mi- ity of life in long-term adult survivors of pediatric solid 25. Centers for Disease Control and Prevention (CDC).
chalek AM. Factors that influence the further survival tumors. Int J Cancer. 1999;(suppl 12):25-31. National Health Interview Survey. US Dept of Hu-
of patients who survive for five years after the diag- 14. Garre ML. Health status of long-term survivors man Health Services, Centers for Disease Control and
nosis of cancer in childhood or adolescence. Med Pe- after cancer in childhood. Am J Pediatr Hematol On- Prevention. Atlanta, Ga: CDC; 1993.
diatr Oncol. 1994;22:91-96. col. 1994;16:143-152. 26. Centers for Disease Control and Prevention (CDC).
2. Green DM, Hyland A, Chung CS, Zevon MA, Hall 15. Feeny D, Furlong W, Barr RD, Torrance GW, Behavioral Risk Factor Surveillance System Survey
BC. Cancer and cardiac mortality among 15-year sur- Rosenbaum P, Weitzman S. A comprehensive multi- Questionnaire. US Dept of Health and Human Ser-
vivors of cancer diagnosed during childhood or ado- attribute system for classifying the health status of sur- vices, Centers for Disease Control and Prevention. At-
lescence. J Clin Oncol. 1999;17:3207-3215. vivors of childhood cancer. J Clin Oncol. 1992;10: lanta, Ga: CDC; 1993.
3. Hawkins MM, Kingston JE, Kinnier Wilson LM. Late 923-928. 27. Diggle PJ, Heagerty P, Liang KY, Zeger SL. Analy-
deaths after treatment for childhood cancer. Arch Dis 16. Zeltzer LK, Chen E, Weiss R, et al. Comparison sis of Longitudinal Data. 2nd ed. Oxford, England: Ox-
Child. 1990;65:1356-1363. of psychologic outcome in adult survivors of child- ford University Press; 2002.
4. Hudson MM, Jones D, Boyett J, Sharp GB, Pui CH. hood acute lymphoblastic leukemia versus sibling con- 28. Hankin BL, Abramson LY. Development of gen-
Late mortality of long-term survivors of childhood can- trols. J Clin Oncol. 1997;15:547-556. der differences in depression: description and pos-
cer. J Clin Oncol. 1997;15:2205-2213. 17. Elkin TD, Phipps S, Mulhern RK, Fairclough D. Psy- sible explanations. Ann Med. 1999;31:372-379.
5. Hudson MM, Poquette CA, Lee J, et al. Increased chological functioning of adolescent and young adult 29. Ettner SL. New evidence on the relationship be-
mortality after successful treatment for Hodgkin’s dis- survivors of pediatric malignancy. Med Pediatr On- tween income and health. J Health Econ. 1996;15:
ease. J Clin Oncol. 1998;16:3592-3600. col. 1997;29:582-588. 67-85.
6. Mertens AC, Yasui Y, Neglia JP, et al. Late mor- 18. Chen E. Pathways linking treatment intensity and 30. Rourke MT, Stuber ML, Hobbie WL, Kazak AE.
tality experience in five-year survivors of childhood and psychosocial outcomes among adult survivors of child- Posttraumatic stress disorder. J Pediatr Oncol Nurs.
adolescent cancer. J Clin Oncol. 2001;19:3163- hood leukemia. J Health Psych. 1998;3:23-28. 1999;16:126-135.
3172. 19. Zebrack BJ, Zeltzer LK, Whitton J, et al. Psycho- 31. Kadan-Lottick NS, Robison LL, Gurney JG, et al.
7. Nicholson HS, Fears TR, Byrne J. Death during adult- logical outcomes in long-term survivors of childhood Childhood cancer survivors’ knowledge about their past
hood in survivors of childhood and adolescent can- leukemia, Hodgkin’s disease, and non-Hodgkin’s lym- diagnosis and treatment: Childhood Cancer Survivor
cer. Cancer. 1994;73:3094-3102. phoma. Pediatrics. 2002;110(1 pt 1):42-52. Study. JAMA. 2002;287:1832-1839.
8. Robertson CM, Hawkins MM, Kingston JE. Late 20. Rourke MT, Hobbie W, Kazak A. Posttraumatic 32. Sahler OJ, Roghmann KJ, Carpenter PJ, et al. Sib-
deaths and survival after childhood cancer: implica- stress in young adult survivors of childhood cancer ling adaptation to childhood cancer collaborative study.
tions for cure. BMJ. 1994;309:162-166. (poster). Paper presented at: 7th International Con- J Dev Behav Pediatr. 1994;15:353-366.
9. Dreyer ZE, Blatt J, Bleyer A. Late effects of child- ference on Long-Term Complications of Treatment of 33. Sloper P, While D. Risk factors in the adjustment
hood cancer and its treatment. In: Pizzo PA PD, ed. Children and Adolescents for Cancer, June 2002; Nia- of siblings of children with cancer. J Child Psychol Psy-
Principles and Practice of Pediatric Oncology. 4th ed. gara-on-the-Lake, Ontario, Canada. chiatry. 1996;37:597-607.
Philadelphia, Pa: Lippincott Williams & Wilkins; 2002: 21. Hobbie WL, Stuber M, Meeske K, et al. Symp- 34. Horwitz WA, Kazak AE. Family adaptation to child-
1431-1461. toms of posttraumatic stress in young adult survivors hood cancer: sibling and family system variables. J Clin
10. Gurney JG, Kadan-Lottick NS, Packer RJ, et al. of childhood cancer. J Clin Oncol. 2000;18:4060- Child Psychol. 1990;4:462-483.
Endocrine and cardiovascular late effects among 4066. 35. Dolgin MJ, Blumensohn R, Mulhern RK, et al. Sib-
adult survivors of childhood brain tumors: Child- 22. Robison LL, Mertens AC, Boice JD, et al. Study ling adaptation to childhood cancer collaborative study:
hood Cancer Survivor Study. Cancer. 2003;97:663- design and cohort characteristics of the Childhood Can- cross-cultural aspects. J Psychosoc Oncol. 1997;15:
673. cer Survivor Study. Med Pediatr Oncol. 2002;38:229- 1-14.
11. Marina N. Long-term survivors of childhood can- 239. 36. Barbarin OA, Sargent JR, Sahler OJZ, et al. Sib-
cer: the medical consequences of cure. Pediatr Clin 23. Derogatis L. Brief Symptom Inventory (BSI) 18: ling adaptation to childhood cancer collaborative study.
North Am. 1997;44:1021-1042. Administration, Scoring, and Procedures Manual. J Psychosoc Oncol. 1995;13:2-20.
12. Anderson DM, Renniem KM, Ziegler RS, Neglia JP, Minneapolis, Minn: National Computer Systems; 37. Havermans T, Eiser C. Siblings of a child with can-
Robison LL, Gurney JG. Medical and neurocognitive late 2000. cer. Child Care Health Dev. 1994;20:309-322.
effects among survivors of childhood central nervous 24. Zabora J, BrintzenhofeSzoc K, Jacobsen P, et al. 38. Sahler OJ, Roghmann KJ, Mulhern RK, et al. Sib-
system tumors. Cancer. 2001;92:2709-2719. A new psychosocial screening instrument for use with ling Adaptation to Childhood Cancer Collaborative
13. Crom DB. Health Status and health related qual- cancer patients. Psychosomatics. 2001;42:241-246. Study. J Dev Behav Pediatr. 1997;18:233-243.

1592 JAMA, September 24, 2003—Vol 290, No. 12 (Reprinted) ©2003 American Medical Association. All rights reserved.

Downloaded From: http://jama.jamanetwork.com/ by a Nanyang Technological University User on 05/19/2015