Positive and Negative Life Changes Experienced by Survivors of Non-Hodgkin’s Lymphoma
Keith M. Bellizzi, Ph.D., MPH
Office of Cancer Survivorship, Division of Cancer Control and Population Sciences, National Cancer Institute,
Melissa Farmer Miller, Ph.D., MPH
Office of Cancer Survivorship, Division of Cancer Control and Population Sciences and Cancer Prevention
Fellowship Program, Division of Cancer Prevention, National Cancer Institute, Bethesda, Maryland
Neeraj K. Arora, Ph.D.
Outcomes Research Branch, Applied Research Program, Division of Cancer Control and Population Sciences,
National Cancer Institute, Bethesda, Maryland
Julia H. Rowland, Ph.D.
Office of Cancer Survivorship, Division of Cancer Control and Population Sciences, National Cancer Institute,
Background: The impact of cancer on adult survivors of
aggressive non-Hodgkin’s Lymphoma (NHL) is understu-
died. Purpose: We examined positive and negative life
changes (health behaviors, relationships, financial situation)
experienced by survivors of NHL and their association with
physical and mental function. Methods: Using the Los
Angeles County Cancer Surveillance Program, 744 question-
naires were mailed to adult survivors of NHL: 308 provided
complete data for analyses (M age ¼ 59.8, SD ¼ 14.9).
Results: Perceptions of positive and negative life changes
were common in our sample, with 77.9% of NHL survivors
reporting at least one positive change and 78.6% reporting
at least one negative change. Cancer had the greatest positive
change on relationships and the most negative change on sur-
vivors’ financial situation. There was an equal distribution of
survivors classified as having experienced positive change and
negative change on health behaviors. Regardless of whether
positive and negative life change were entered into separate
regression models or the same model, an increase in negative
life change in each of the domains was significantly associa-
ted with a decrease in physical and mental functioning. Posi-
tive change was significantly associated only with physical
functioning when examining overall change (p ¼ .018) and
health behaviors (p ¼ .013), and the inclusion of negative
change attenuated these associations. Conclusions: In design-
ing interventions to improve the mental and physical function
of NHL survivors, the greatest benefit may likely be achieved
by reducing the negative effects of cancer. Perhaps positive
life changes are related in more specific ways to other
indexes of adjustment, but our findings failed to show a posi-
tive relationship with mental and physical function.
(Ann Behav Med
Non-Hodgkin’s Lymphoma (NHL) ranks amongst the
top six cancer sites in terms of incidence and the top eight
for mortality for both men and women (1). Since the 1970s,
incidence rates for NHL have increased dramatically, mak-
ing it one of the fastest rising cancers in the United States.
Overall, the 5-year survival rate for NHL is 60% (1).
A large proportion of adult NHL cases are classified as
aggressive in nature (51%). Individuals with aggressive
NHL receive intense treatment, which may include multi-
agent chemotherapy regimens, radiation, and bone mar-
row=stem cell transplantation. NHL and its treatments
can have a major impact on survivors’ lives and quality
of life. Identifying how the impact of cancer and its treat-
ment relates to physical and mental function is critical to
understanding the posttreatment experience of this rapidly
growing yet understudied population.
We know from several studies that cancer and its treat-
ment can negatively affect various life domains of cancer
survivors (2–8). These studies have found cancer survivors
The collection of cancer incidence data used in this study was
supported by the California Department of Health Services
(CADHS) as part of the statewide cancer reporting program man-
dated by California Health and Safety Code Section 103885; the
National Cancer Institutes (NCI) Surveillance, Epidemiology
and End Results Program under contract N01-PC-35139 awarded
to the University of Southern California, and contract N02-PC-
15105 awarded to the Public Health Institute (PHI); and the
Centers for Disease Control and Prevention’s (CDC) National
Program of Cancer Registries, under agreement #U55=CCR
921930-02 awarded to the PHI. The ideas and opinions expressed
herein are those of the author(s) and endorsement by the CADHS,
the NCI, and the CDC or their contractors and subcontractors
neither is intended nor should be inferred.
Reprint Address: K. Bellizzi, Ph.D., MPH, National Cancer
Boulevard, Suite 404, MSC 8336, Bethesda, MD 20892. E-mail:
# 2007 by The Society of Behavioral Medicine.
to experience negative changes in their social relationships
(4), finances (6), and educational and career attainment
(4,6), and they report elevated psychosocial concerns like
fear of recurrence (5). Conversely, emerging evidence sug-
gests that cancer survivors, similar to survivors of other
traumatic life events, also may experience positive life
changes (or posttraumatic growth) as a result of their
illness (9–15). Most notably, cancer survivors report
positive changes in their family relationships, priorities=
direction in life, spirituality, and appreciation for life
(9,14,16–18). Although these studies have significantly con-
tributed to the literature, they are generally restricted to
breast cancer survivors. We know little in terms of positive
and negative life changes reported by adult survivors of
NHL and how these changes affect their physical and men-
tal function. Survival rates for NHL are favorable in many
cases, meaning men and women can expect to live years
after their diagnosis and have sufficient time to ruminate
about and integrate their experience into their lives, which
is one of the theoretical foundations of growth (19).
The extant research in this area suffers from some
important methodological limitations that warrant investi-
gation. With few exceptions (20–22), reports of positive
and negative life changes following cancer and other trau-
matic life events (15,23) have largely been examined inde-
pendent from one another. One reason for this restricted
methodological approach is that existing measures of posi-
tive life change, the Posttraumatic Growth Inventory
(PTGI) (24), the Stress Related Growth Scale (SRGS)
(25), the Perceived Benefits Scale (PBS) (26) and the Benefit
Finding Scale (27,28) only include positive change.
Although the PBS includes negatively worded items in
the scale, these are added to reduce response bias, and
the authors suggest dropping them from analyses. These
measures were developed to focus primarily on positive
changes due to the novelty of this area and the abundance
of evidence examining the negative consequences of trau-
matic events. However, survivors of cancer report the
coexistence of positive and negative changes (7,22).
Concerns persist among researchers over the validity and
inherent biases of instruments that measure one side of
the life change continuum (11,16,29). Over reliance on
instruments that capture only one aspect of the impact of
an adverse life event may cause a false positive response
bias or self-enhancing bias of the attribute being studied
(11,16). The study presented here attempts to address these
concerns by including positive and negative life change
items in one scale and accounting for coexisting negative
and positive change in the analysis. An additional
limitation of existing instruments is that they fail to capture
other potentially salient domains of life change identified
in related literature. For example, health behaviors and
financial situation were identified in studies of people
living breast cancer and AIDS, respectively (6,30). We
included items in the life change scale to account for these
Several studies exist that examine the association
between positive life change and various indices of
health-related quality of life (HRQOL), including mea-
sures of depression, general distress, anxiety, positive
and negative affect, and general well-being. By and large,
findings are inconsistent with some showing a positive
association with measures of HRQOL (7,27,31,32), others
a negative association (14), and still others no association
at all (9,33). For two thorough reviews of this literature,
see Stanton et al. (18) and Zoellner and Maercher (34).
These mixed findings could be explained by differences
in measures used to assess growth and HRQOL. For
instance, Tomich and Helgeson’s (27) study of 164 breast
cancer survivors assessed growth with a one-item ques-
tion, their (2004) study of 364 women diagnosed with
stage 1, 2, or 3 breast cancer used a 17-item global
measure, and Carpenter and colleagues’ (7) study of
breast cancer survivors used several measures of well-
being and an one open-ended question to capture positive
changes after cancer diagnosis. An additional limitation
of much of the previous work is the lack of power to
detect a significant difference (9,33). Despite several
cross-sectional and longitudinal attempts at assessing the
relationship between posttraumatic growth and quality
of life, clear evidence supporting one direction versus
the other is arguably lacking. The research reported here
extends previous work by concurrently examining positive
and negative life change and other indices of HRQOL,
namely, physical and mental function in a large popu-
lation-based sample of NHL survivors.
Although not the primary focus of this article, we also
examined theoretically and empirically linked socio-
demographic, disease-related, and psychological factors
associated with perceptions of positive and negative life
changes. Past research has examined variables, such as
age, gender, marital status, education, stage of disease,
treatment, time since diagnosis, optimism and social sup-
port, to name a few (for a review of this literature related
to cancer, see 18). With few exceptions, most of the corre-
lates that have been examined in previous studies show
inconsistent associations with measures of life changes.
Despite these inconsistencies we rely on previous theory
to inform our hypotheses. The theory of cognitive adap-
tation suggests that, in the face of a stressful event such
as cancer, individuals attempt to control their disease by
the use of optimistic and positive thinking (31,35). Tedeschi
and Calhoun’s model of growth suggest that optimism
might be part of the pathway to positive life changes
(10,35). Schaefer and Moos’s model of life crises and per-
sonal growth and Tedeschi and Calhoun’s model of post-
traumatic growth posit that social support plays a role in
facilitating positive life changes (36,37). This study exam-
ines some of the more commonly examined correlates of
life change, including optimism and social support, using
a measure of change that allows for both positive and
Volume 34, Number 2, 2007
Life Change and HRQOL189
There were three specific aims to this study. The first
was to describe the proportion of NHL survivors experi-
encing positive and negative change after cancer. Based
on past research, we hypothesized that survivors of NHL
will report the co-occurrence of positive and negative life
changes after cancer. The second aim was to describe the
disease-related factors, psychosocial factors (i.e., optimism
and social support) and positive and negative life change.
Based on theory, we hypothesized those NHL survivors
who report higher optimism and more social support will
report more positive life change and less negative life
change than those with lower scores on optimism and
social support. Our third aim was to examine the relation-
ship between positive and negative change on survivors’
physical and mental function. Due to the inconsistencies
in the literature regarding the association between life
change and mental and physical and mental health, this
aim is exploratory.
Participants and Procedures
Adult survivors (2 to 5 years since diagnosis) of
aggressive NHL (i.e., intermediate and high grades of
NHL) diagnosed between June 1, 1998, and August 31,
2001, were selected from the Los Angeles Cancer Surveil-
lance Program. Eligibility criteria for the participants
included (a) age 20 years or older, (b) alive at the time of
the study, (c) primary cancer diagnosis of NHL, and (d)
ability to read and write in English. Eligibility was also lim-
ited to individuals classified in the registry database as
either African American or White (including both non-
Latino or non-Spanish Surnamed Whites) and Latinos
(or Spanish Surnamed Whites). In total, 1,025 cases were
identified by the registry (see Figure 1).
Of the 1,025 selected cases, 744 were eligible after elim-
inating those found to be deceased (n ¼ 109), incompetent
or too ill (n ¼ 36), unable to understand English (n ¼ 80),
having had another cancer (43), or otherwise ineligible
(n ¼ 13; not Los Angeles county resident, prisoner, mis-
diagnosis). Between April 2003 and August 2003, eligible
cases were sent an informed consent form, an introductory
letter explaining the study, and the self-report question-
naire for them to complete. A postage-paid business reply
envelope was included for return of the questionnaire and
informed consent forms.In total there were 408 completed
questionnaires (response rate of 54.8% of those considered
eligible). Twenty-four percent (181 patients) of the 744 eli-
gible were unable to be located despite extensive follow-up
and tracing efforts. Of the 155 patients who refused, about
half were considered to be ‘‘hard’’ refusers (i.e., they stated
that they did not want to participate), whereas the other
Flowchart describing recruitment of non-Hodgkin’s Lymphoma survivors for the study.
190 Bellizzi et al.
Annals of Behavioral Medicine
half were classified as ‘‘soft’’ refusers (i.e., they may have
promised to send in their questionnaire but never did,
never answered the phone or responded to messages, or
gave noncommittal responses but never completed the
questionnaire). The length of the questionnaire and lack
of stamina of the patients were cited as the most common
reasons for nonresponse. A further indication of the dif-
ficulty posed in completing the survey is reflected by the
fact that in 89 (21.8%) of the 408 cases, the survey was
completed by telephone as a last resort; however, the
method of completion (phone vs. mail) was not signi-
ficantly associated with the patient’s age, gender, race=
ethnicity, histology, HIV status, or year of diagnosis.
Respondents who chose a phone interview were not given
the full set of measures (excluded the Short Form–36
[SF-36] and other quality-of-life measures) because of the
length of administration. As a result, our sample was based
on the 319 respondents who completed the questionnaire
by mail. However, we excluded 11 respondents that were
missing information for sociodemographic or psychosocial
variables. The final sample for analyses included 308
respondents. A multivariate logistic regression model was
used to determine if response was independently predicted
by respondent characteristics. Two variables were associa-
ted with nonresponse: male gender and Latino race=ethni-
ethnicity. Among the eligible cases, the adjusted odds ratio
for non-responders was 1.39 (95% C.L. 1.03–1.86) for
male gender and 1.68 (95% C.L. 1.21–2.34) for Latino
race=ethnicity. The poorer response among Latinos was
likely because the interview was not available in Spanish.
This study was approved by the Institutional Review
Board at the University of Southern California.
information included gender, age, race=ethnicity, marital
status, income, and
characteristics included treatment
diagnosis.Time since diagnosis
subtracting years between age at interview and reported
age at initial diagnosis. Sociodemographic and treatment
information were obtained from patient surveys and date
of diagnosis was obtained through the SEER registry
Life impact. To identify the positive and negative
impact of cancer, participants completed a 19-item Life
Impact Scale. Individuals were asked, ‘‘Looking back,
since the time you were first diagnosed with lymphoma,
how much of an impact has cancer and its treatments
had on the following areas of your life?’’ Response
choices were based on a 6-point scale, ranging 0 (does not
apply), 1 (very negative), 2 (somewhat negative), 3 (no
impact), 4 (somewhat positive), and 5 (very positive). Items
were selected based on previous literature on positive and
negative changes following trauma, including cancer,
which identifies areas of life shown to be important to
individuals (6,14,24,31). Nine of the 19 items were
adapted from a study by Ganz et al., to assess perceived
positive and negative impact of cancer on areas of
(relationships, financial situation, health behaviors) that
have been found to be important in survivors of cancer
and other traumatic events (6,24,30). Aim 1—namely, to
describe the proportion of positive and negative life
impact in survivors of NHL—used all 19 items from the
Life Impact Scale.
A principal component analysis followed by Promax
rotation was conducted on the 19-item Life Impact Scale,
yielding a six-factor solution with eigenvalues greater than
1. These six factors were not interpretable, thus we forced
a five-factor solution. In this solution, we had five items
that had loadings of 0.4 or more on two factors and were
subsequently dropped from analyses (education, work life,
ability to date, sex life, and social activities). This process
resulted in the retention of 14 items, which yielded a five-
factor solution of which three factors (10 items) were
interpretable. The three-factor solution had factor load-
ings ranging from .40 to .89 (Table 1). These factors
accounted for 45.0% of the common variance in our life
impact data. To establish internal consistency reliability
we assessed the Cronbach’s alpha for the three factors
along with the intercorrelations among the various scales
that constitute the Life Impact Scale. Cronbach’s alpha
estimates by scale were relationships=life outlook, .74;
health behaviors, .72; and financial situation, .61. Pearson
product-moment correlations among the positive life
change factor scales ranged r ¼ .20 to r ¼ .33, whereas
the correlations among the negative life change factor
scales ranged r ¼ ?.06 to r ¼ ?.11, indicating modest
overlap but distinct contributions from each factor scale
Because we were interested in assessing both positive
and negative changes, we created subscales reflecting the
number and magnitude of self-reported positive (ratings
of 4 or 5) and negative (ratings of 1 or 2) changes overall
and for each domain (relationships=outlook, health behav-
iors, financial situation). For the positive score, a rating of
4 for somewhat positive was given a score of 1, a rating of 5
for very positive was given a score of 2, and all other ratings
(0–3) were given a score of 0 to minimize loss of data.
Including no change in the score is consistent with the scor-
ing of other measures of growth (24,25). Similarly for the
negative score, a rating of 2 for somewhat negative was
given a score of 1, a rating of 1 for very negative was given
a score of 2, and all other ratings (0, 3–5) were given a score
of 0. These positive and negative scores (using the 10 inter-
pretable items from the principal component analysis) were
summed overall and for each domain and used to address
Aims 2 and 3.
Volume 34, Number 2, 2007
Life Change and HRQOL191
HRQOL. The physical health (PCS) and mental health
(MCS) summary scores of the SF-36 were used to measure
HRQOL (38,39). We first computed the scores for the eight
subscalesof the SF-36;
transformed to a 0 to 100 scale such that a score of 0
represented lowest level of functioning and 100 represented
optimal functioning. Final PCS and MCS summary scores
were then constructed based on the 1999 U.S. population
norms; these two scores had a mean value of 50
representing the U.S. population norms and a standard
deviation of 10 (39). To evaluate the association between
life change and physical and mental function, we restricted
analyses to the two component summary scores to reduce
the number of analyses. The SF-36 is a well-established
instrument and has shown good reliability and validity in
measuring physical and mental function in cancer survivors
and persons with other medical conditions (40,41).
thesescores were linearly
Optimism and social support. The Life Orientation
Test—Revised was used to measure optimism (42). This
eight-item scale includes questions such as ‘‘In uncertain
times, I usually expect the best’’ and ‘‘If something can
go wrong for me, it will.’’ The scale has exhibited good
populations (43). Social support was measured using a
shortened version (6) (12 items) of the MOS Social
Support Measure (19 items) (44). Respondents were
asked, ‘‘How often is each of the following kinds of
support available to you if you need it?’’ Items include
‘‘someone to take you to the doctor’’ and ‘‘someone who
understands your problems.’’ All items are scored on a
5-point Likert scale from none of the time to all of the
time. An overall social support scale score was created by
taking the mean of the scores on the 12 individual items.
We then linearly transformed the social support score to
a 0–100 scale, with a higher score indicating higher
perceived available social support.
use with chronicallyill
Summary measures of positive and negative life change
included mean score and standard error. For each individual
Pearson Product-Moment Correlations Among the Life Impact Scores
Pos. health behavior
Neg. health behavior
Note. The positive and negative overall scores include all 19 items of the Life Impact Scale. Pos. ¼ positive; Neg. ¼ negative.
?p ¼ .05.??p ¼ .01.
Oblique Rotated Factor Pattern From Principal Component Analysis of the Life Impact Data
Relationships=Life Outlook Health Behaviors Financial Situation
Relationship with spouse=partner
Relationship with children
Relationship with family and friends
Enjoyment in life
Note. n ¼ 275. The three factors account for 45.0% of the common variance in our life change data. Items removed from the scale, include
education, work life, ability to date, sex life, alcohol consumption, smoking, ability and desire to have children, and social activities. Factor
loadings used to interpret the meaning of each factor are highlighted in bold.
192 Bellizzi et al.
Annals of Behavioral Medicine
item in the 19-item life change scale, we reported the
proportion of NHL survivors reporting positive change,
negative change, no impact, and does not apply. We used
multivariable linear regression to identify correlates of
positive and negative life change. From these models, we
estimated adjusted mean scores and standard errors for
positive and negative change for each level of a categorical
variable, and we report the b-coefficient and standard errors
for continuous variables. In addition, linear regression mod-
els were estimated with mental and physical functioning
scores as dependent variables and life change (overall and
for each domain) as independent variables. The following
covariates were associated with either mental or physical
functioning in bivariate analysis and were therefore included
to control for confounding: gender, age, race=ethnicity,
marital status, education, social support, optimism, time
since diagnosis, and treatment. We report the results from
multivariable regression analysis where positive and negative
life change were considered separately and were included in
the same regression model. Multicollinearity among predic-
tors in our regression analysis was tested using variance
inflation factors and all were less than 2.0.
Sociodemographic and Disease Characteristics
of NHL Survivors
The finalsamplefor the studyconsistedof308NHLsur-
vivors with aggressive histology. Respondents were, on aver-
age, 60 years of age ranging from 23 to 85 years. Almost two
thirds (64.9%) of the respondents were living with a spouse
or other partner. Thirty percent self-identified as Hispanic.
Gender was evenly distributed, with 48.7% female. Over
one third (37.7%) reported having completed college and=or
had chemotherapy (94.2%) as part of their treatment. The
majority of NHL survivors never had a recurrence (82.7%),
whereas an even larger proportion was self-reportedly in
remission (89.1%). The mean (SE) time since diagnosis for
the respondents was 3.5 (0.1) years.
Impact of Cancer on the Lives of NHL Survivors
In this sample, 77.9% of NHL survivors reported at
least one positive change, and 78.6% reported at least
one negative change. Only 12.4% of the sample reported
no change, either positive or negative as a result of their
cancer. Table 3 reports the proportion of NHL survivors
reporting positive change, negative change, and no
impact=does not apply for the 19 items of the Life Impact
Scale, listed in ranked order from most positive impact to
least positive impact. Of note, the item with the highest
proportion of respondents endorsing positive change
was in relationships with family members and friends
(43.4%), followed by positive changes in religious or spiri-
tual beliefs (42.2%). Items with a high proportion of
respondents reporting negative change included changes
in work life or career (34.2%), financial situation
(33.9%), and sex life (32.5%). There were a large pro-
portion of respondents who reported no change or does
not apply to several of the items, including education and
The Proportion of non-Hodgkin’s Lymphoma Survivors Reporting Positive Change, Negative Change, No Impact, and Does Not Apply
Does Not Apply
Your relationship with other family members and friends
Your religious or spiritual beliefsa
Your relationship with your children
Your ability to enjoy life
Your relationship with your spouse=partner
Your exercise activitiesa
Your work life or careera
Your participation in social activitiesa
Your retirement plansa
Your education plans
Your financial situationa
Your smoking of tobacco products
Your ability to get or retain health, life, or disability insurancea
Your sex lifea
Your alcohol consumption
Your ability to date people
Your desire to have children
Your ability to have children
aThis item was adapted from the Ganz et al. 2002 study (6).
Volume 34, Number 2, 2007
Life Change and HRQOL193
work life, smoking of tobacco products, and the questions
related to having children, which is likely due to the high
mean age of respondents in our sample.
NHL survivors reported both positive (M ¼ 3.95,
SE ¼ 0.22) and negative (M ¼ 3.30, SE ¼ 0.23) life change
since the time they were first diagnosed with lymphoma,
and, on average, the magnitude of positive and negative
change was equivalent (p ¼ .223). Similarly, they reported
a comparable amount of positive change (M ¼ 0.60,
SE ¼ 0.06)
(M ¼ 0.63, SE ¼ 0.06, p ¼ .345). However, respondents
reported more negative than positive change in their finan-
cial situation (M ¼ 1.24, 0.28; SE ¼ 0.09, 0.05, respect-
ively), and more positive than negative change in their
relationships=outlook (M ¼ 2.76, 0.76; SE ¼ 0.16, 0.08,
respectively; ps < .001).
in healthbehavior asnegativechange
Differences by Sociodemographic and
Using linear regression, we calculated adjusted mean
scores to examine the association between sociodemo-
graphic, disease-related, and psychosocial factors with both
positive and negative life change scores (Table 4). Findings
show older age was independently associated with lower
negative life change scores, overall and in the three
domains (ps < .05). Older age was also associated with
lower positive life change in overall life change (p < .05).
In addition, greater education was associated with higher
positive life change in health behaviors (p < .05) and hav-
ing a significant other was associated with lower positive
change in financial situation. Available social support
was associated with higher positive life change and lower
Adjusted Mean Scores of Life Change by Demographic, Clinical, and Psychosocial Characteristics of Sample
Positive Negative PositiveNegativePositiveNegative Positive Negative
Total mean (SE)
Adjusted mean (SE)b
Years from diagnosis 0.12 (0.26)
3.95 (0.22) 3.30 (0.23)2.76 (0.16) 0.76 (0.08) 0.60 (0.06)0.63 (0.06) 0.28 (0.05)1.24 (0.09)
0.01 (0.07) ? 0.18 (0.06)
–0.07 (0.01) ? 0.01 (0.01)
–0.01 (0.01) ? 0.01 (0.01) ? 0.01 (0.01) ? 0.01 (0.01) ? 0.02 (0.01)
0.03 (0.05) ? 0.08 (0.02)
0.06 (0.19) ? 0.02 (0.09)
0.01 (0.02) ? 0.02 (0.01) ? 0.02 (0.01) ? 0.06 (0.02)
0.03 (0.07) ? 0.08 (0.06)0.01 (0.25)0.04 (0.05)0.09 (0.11)
Note. Values in bold indicate p-value <0.05 from overall F-test, with the exception of education and social support where the p-value from a
trend test was used.
aValues are adjusted for all other variables listed in the table.
cOr living as married.
dFrom a shortened version (12 items) of the MOS Social Support Measure (19 items). Scores in quintiles.
eFrom Life Orientation Test - Revised.
194 Bellizzi et al.
Annals of Behavioral Medicine
negative life change scores overall and in the relationship
domain. Higher optimism was associated with lower
negative life change overall and in relationships and finan-
cial situation (p < .05).
Modeling the Relationship Between Life Change
and Physical and Mental Functioning
We modeled the relationship between physical and
mental functioning and positive and negative life change,
adjusted for sociodemographic, disease, and psychosocial
characteristics (Table 5). Findings show that regardless of
whether positive and negative life change were entered into
separate regression models or the same model, an increase
in negative life change overall and in each of the three
subdomains was significantly associated with a decrease
in physical and mental functioning. On the other hand,
positive change was significantly associated only with
physical and mental functioning when examining overall
life change and change in health behavior (p ¼ .024 and
.026, respectively), and the inclusion of negative change
attenuated these associations so that they were no longer
The goal of this study was to describe positive and nega-
tive life changes concurrently in adult survivors of NHL and
examine how these life changes relate to physical and mental
functioning. Our results indicate that perceptions of positive
and negative life changes were common in our sample of
NHL survivors, consistent with previous reports of life
change in cancer survivors (7,9,17,28). Supporting our first
hypothesis, NHL survivors in our sample reported the
co-occurrence of positive and negative life change as a result
of their experience, adding to the sparse evidence that both
of these responses coexist in cancer survivors (20).
An important contribution of this study was the meth-
odological approach used to analyze the data from the Life
Impact Scale. In the growth field there is no standard way
of scoring this type of data despite the recognition that cap-
turing both positive and negative life changes is important
(11,16,18,45). To be most consistent with the scales that
measure positive life changes (i.e., PTGI, SRGS, PBS),
we computed a positive score and a separate negative score
overall and for each domain. This allowed us to examine
the independent effects of positive life changes while
adjusting for negative change. Another approach that we
considered was to compute one continuous score consistent
with the scoring of Likert scales. The drawback of this
approach is that aggregating positive and negative change
could obscure the potential importance of reporting both
positive change and negative change in the same domain.
For example, if, following cancer, one becomes estranged
from friends but closer to one’s significant other, then it
is possible we will find ‘‘no change’’ or a washout effect,
if we examine average scores. Without complete represen-
tation of a domain, then the domain becomes a function
of the specific items that make it up. We did, however,
score the items using this latter approach and our findings
were consistent with findings presented in this paper. Both
methods of scoring showed that physical and mental
Simultaneous Regressions Testing Associations Between Life Change and Physical and Mental Health
Physical Component SummaryMental Component Summary
b (SE)t Testp
b (SE) t Testp
< .001< .001
< .001< .001
.76 0.033 (0.71)
< .001< .001
Note. Values are adjusted for gender, age, race=ethnicity, marital status, education, social support, optimism, time since diagnosis and treat-
ment. Positive and negative life change scores are included in one model. Model statistics for life change and physical health: Overall life change,
adj. R2¼ .23, F(18, 278) ¼ 6.19, p < .001; Relationships=outlook life change, adj. R2¼ .19, F(18, 278) ¼ 4.94, p < .001; Health behaviors life
change, adj. R2¼ .28, F(18, 276) ¼ 7.25, p < .001; Financial situation life change, adj. R2¼ .21, F(18, 277) ¼ 5.25, p < .001. Model statistics
for life change and mental health: Overall life change, adj. R2¼ .33, F(18, 278) ¼ 8.96, p < .001; Relationships=outlook life change, adj.
R2¼ .32, F(18, 278) ¼ 8.82, p < .001; Health behaviors life change, adj. R2¼ .28, F(18, 276) ¼ 7.39, p < .001; Financial situation life change,
adj. R2¼ .25, F(18, 277) ¼ 6.37, p < .001.
Volume 34, Number 2, 2007
Life Change and HRQOL 195
functioning decreased significantly with more negative
change but was not associated with positive life change.
The differential effect of NHL on the domains of life
change suggest that cancer does not uniformly affect differ-
ent areas of life offering evidence to support the multidimen-
sionality of this positive life change scale. Overall, we found
that NHL survivors reported an equal proportion of posi-
tive and negative changes after cancer. However, when we
examined the three domains of the life change measure, dif-
ferent patterns emerged. On average, there was a negative
impact in individual’s financial situation, which is likely
the result of the very real strains the cancer experience can
have on a survivor’s employment status, income, health care
expenditures, and retirement plans. Conversely, we found a
predominantly positive impact of cancer on respondents’
relationships, which is consistent with previous research
(9,17). Findings related to the impact of cancer on health
behaviors suggest an equal proportion of positive and nega-
tive change adding to the limited evidence (30) that health
behaviors is a domain that should be included in measures
of life change. In addition, the finding that health behaviors
were not affected predominantly in the positive direction,
as one might expect, suggests the further need for health
promoting interventions and wellness counseling for NHL
survivors. There is some evidence that suggests the posttreat-
ment patient debriefing provides a teachable moment when
cancer survivors are more willing to take an active role in
their posttreatment care (46).
Research suggests physician delivered exercise pro-
motion messages are a powerful catalyst in motivating
exercise behavior change (47–49).Thus, providers of care
to NHL survivors may be in an ideal position to provide
healthy lifestyle counsel and guidance to their patients
during the course of, and after, cancer treatment.
Consistent with previous work in this area (7,20,43,50),
older age was associated with respondents being less likely
to report that cancer had a positive change on their lives. It
is interesting to note that older age was also associated with
lower negative life change, suggesting that older cancer sur-
vivors are less affected by their cancer experience. An alter-
native explanation might be that older cancer survivors are
dealing with other more salient normative life events asso-
ciated with the aging process and the impact of cancer is
thus perceived as a less significant event. Previous work
examining the impact of cancer on the lives of prostate can-
cer survivors shows that some older men with prostate can-
cer who reported no impact of the disease were dealing
with vision losses associated with aging, recent loss of a
loved one, and other more traumatic life events (50). There
are also indications in other research that some survivors of
prostate cancer feel other life experiences have had more
impact on their lives than has cancer (51). Thus, when older
cancer survivors are asked about the impact cancer has had
on their lives, we should consider how cancer is assimilated
in relation to other major life events and normative life
Other correlates of life change found in this study
include education, available social support, and optimism.
We found partial support for our second hypothesis,
namely, higher available social support scores were associa-
ted with higher levels of positive life changes overall and in
relationships, similar to prior research (25,52). In addition,
we found higher social support was related to lower levels
of negative life change in relationships and financial situ-
ation. Although optimists are more likely to report lower
levels of negative life change, being optimistic was not asso-
ciated with an increased likelihood of reporting positive
changes. Although being optimistic might be protective
against negative consequences cancer, it may not necessar-
ily equate to being influential in similar, but also very dif-
ferent, measures of positive life changes. Higher education
was associated with respondents reporting more positive
life change in health behaviors, which is similar to previous
findings that found a relationship between higher edu-
cation and positive life changes (9,12). It is possible that
positive changes in health behaviors are more likely for
individuals with higher education as a result of personal
as well as physical resources.
Our findings do not support the premise that percep-
tions of positive life change measured using the Life Impact
Scale are associated with enhanced physical and mental
functioning, as measured by the mental and physical
component summary scores of the SF-36. Because of the
heterogeneity of measures and population characteristics
across other studies, it is inherently difficult to make direct
comparisons with our study. To our knowledge, only one
other study of positive life change has used the SF-36 as
a measure of adjustment, and the investigators of that
study observed a modest association (r ¼ .17) between a
one-item benefit finding question and physical function
(27). The inconsistency between their study and ours could
be explained by the two different measures of positive life
change used. Conceptually, our findings are consistent with
research that shows no relationship with other measures of
adjustment (9,33) but inconsistent with other studies that
show a positive association (7,20,27,32). Although it might
seem intuitive that growth should be associated with physi-
cal and mental functioning, we need to consider that
growth occurs in the context of highly stressful events
(i.e., cancer), which means it likely co-occurs with distress
(53), thus there may be a wash in terms of the impact on
physical and mental function.
Despite inconsistent findings, our study is a unique
contribution to this discourse because it is one of the first
studies to examine positive life changes and mental and
physical function in the context of negative change coexist-
ing within each domain. This approach addresses previous
positive response bias concerns (11,16) and suggests that in
designing interventions to improve the physical and mental
function of NHL survivors, the greatest benefit might
be achieved by reducing the negative effect of cancer. Find-
ings from this study do not provide evidence to support
196Bellizzi et al.
Annals of Behavioral Medicine
interventions that encourage positive life changes as a
means for improving mental and physical function. It is
possible that measures of overall adjustment or general
function, such as the SF-36, do not capture the potential
benefit that positive life change has on individual’s lives.
Longitudinal studies of growth are needed and should
explore other indexes of adjustment that might be related
in a more specific way, such as resilience or preparedness
for future health challenges, health competence, satisfac-
tion with life and relationships, identity development,
expectations, body image, and self-image.
This study has certain limitations that need to be
acknowledged. First, the cross-sectional nature of our data
restricts our ability to make causal associations. However,
recognizing the challenges of longitudinal research, Calhoun
and Tedeschi (54) contended that cross-sectional studies are
still useful in this field, especially when looking at new rela-
tionships.Second, our measure of lifechangehas several lim-
itationsthat require enumeration.
evaluation of the Life Impact Scale is limited to our study.
The Life Impact Scale is not inclusive with respect to all the
potential life domains important to survivors of cancer.
Nevertheless, our findings did include two new domains—
health behaviors and financial situation—which could be
considered in future research in this field. Although the three
financial situation items in the Life Impact Scale are intuit-
ively related, the low internal consistency (a ¼ .61) may sug-
gest they are not capturing a unified construct. Third, the
generalizability of this study is limited to the demographic
and disease-related characteristics of our sample. As men-
tioned previously, there was a bias toward nonparticipation
by men and Hispanics. Despite few differences between
responders and nonresponders, our sample likely consisted
of a healthier population as demonstrated by lack of stamina
being cited as one of the primary reasons for nonresponse.
Last, our data relied on patient self-report, so inherent in
our findings are the biases that go along with this method
of data collection, including recall bias.
This study contributes to a growing body of research
suggesting survivors of NHL, similar to survivors of other
cancers and traumatic life events, report the coexistence of
positive and negative life change. This population-based
study is the first to examine both positive and negative life
change concurrently in survivors of NHL and offers a more
comprehensive understanding between life change and
physical and mental health. By including both positive
and negative items in the Life Impact Scale, this study
attempted to address the positive response bias criticism
of previous work. Results from our principal component
analysis suggest two new life change domains worthy of
future investigation. Although the association between
positive life change and mental and physical function was
not significant, we suggest future studies might examine
other aspects of adjustment that might have more specific
relationships with different domains of positive life change.
Last, although we cannot draw causal associations from
this cross-sectional study, it would be hard to ignore the
potential benefit that could be obtained in physical and
mental function by designing an intervention to reduce
the negative life impact of cancer given the strong and
consistent association in this study.
(1) American Cancer Society: American Cancer Society Cancer
Facts and Figures. Atlanta, GA: American Cancer Society,
(2) Devlin J, Maquire P, Phillips P: Psychological problems
associated with diagnosis and treatment of lymphomas. II.
Prospective study. British Medical Journal. 1987, 295:
(3) Carter R, Carter C: Individual and marital adjustment
in spouse pairs subsequent to mastectomy. The American
Journal of Family Therapy. 1993, 21:291–300.
(4) Glanz K, Lerman C: Psychosocial impact of breast cancer:
A critical review. Annals of Behavioral Medicine. 1992,
(5) Mehta SS, Lubeck DP, Pasta DJ, Litwin MS: Fear of
recurrence in patients undergoing definitive treatment for
prostate cancer: Results from CaPSURE. The Journal of
Urology. 2003, 170:1931–1933.
(6) Ganz PA, Desmond KA, Leedham B, et al.: Quality of life
in long-term, disease-free survivors of breast cancer: A
follow-up study. Journal of the National Cancer Institute.
(7) Carpenter JS, Brockopp DY, Andrykowski MA: Self-
transformation as a factor in the self-esteem and well-being
of breast cancer survivors. Journal of Advanced Nursing.
(8) Zabora J, BrintzenhofeSzoc K, Curbow B, Hooker C,
Piantadosi S: The prevalence of psychological distress by
cancer site. Psycho-Oncology. 2001, 10:19–28.
(9) Cordova MJ, Cunningham LLC, Carlson CR, Andry-
kowski MA. Posttraumatic growth following breast cancer:
A controlled comparison study. Health Psychology. 2001,
(10) Tedeschi RG, Calhoun LG: Trauma & Transformation:
Growing in the Aftermath of Suffering. Thousand Oaks,
CA: Sage, 1995, p. 163.
(11) Park CL, Lechner SC: Measurement issues in assessing
growth following stressful life experiences. In Calhoun
LG, Tedeschi RG (eds), Handbook of Posttraumatic
Growth: Research and Practice. Mahwah, NJ: Lawrence
Erlbaum Associates, 2006.
(12) Sears SR, Stanton AL, Danoff-Burg S: The yellow brick
road and the emerald city: Benefit finding, positive reap-
praisal coping and posttraumatic growth in women with
early-stage breast cancer.
(13) Andrykowski MA, Cordova MJ, Studts JL, Miller TW:
Posttraumatic stress disorder after treatment for breast
cancer: Prevalence of diagnosis and use of the PTSD Check-
list–Civilian Version (PLC-C) as a screening instrument.
Journal of Consulting & Clinical Psychology. 1998, 66:
Volume 34, Number 2, 2007
Life Change and HRQOL 197
(14) Tomich PL, Helgeson VS: Is finding something good in the
bad always good? Benefit finding among women with breast
cancer. Health Psychology. 2004, 23:16–23.
(15) Frazier P, Tashiro T, Berman M, Steger M, Long J: Corre-
lates of levels and patterns of positive life changes following
sexual assault. Journal of Consulting & Clinical Psychology.
(16) Linley P, Joseph S: Positive change following trauma and
adversity: A review. Journal of Traumatic Stress. 2004,
(17) Bellizzi KM, Blank TO: Predicting posttraumatic growth
(18) Stanton AL, Bower JE, Low CA: Posttraumatic growth
after cancer. In: Calhoun LG, Tedeschi RG (eds), Hand-
book of Posttraumatic Growth: Research and Practice.
Mahwah, NJ: Lawrence Erlbaum Associates, 2006.
(19) Tedeschi RG, Park CL, Calhoun LG: Posttraumatic
Growth: Positive Changes in the Aftermath of Crisis.
Mahwah, NJ: Lawrence Erlbaum Associates, 1998, p. 258.
(20) Bower JE, Meyerowitz BE, Desmond KA, et al.: Percep-
tions of positive meaning and vulnerability following breast
cancer: Predictors and outcomes among long-term breast
cancer survivors. Annals of Behavioral Medicine. 2005,
(21) Klauer T, Ferring D, Filipp S-H: ‘‘Still stable after all
this...?’’: Temporal comparison in coping with severe and
chronic illness. International Journal of Behavioral Develop-
ment. 1998, 22:339–355.
(22) Curbow B, Somerfield MR, Baker F, Wingard JR, Legro
MW: Personal changes,
psychological adjustment to bone marrow transplantation.
Journal of Behavioral Medicine. 1993, 16:423–443.
(23) Frazier P, Conlon A, Glaser T: Positive and negative life
changes following sexual assault. Journal of Consulting &
Clinical Psychology. 2001, 69:1048–1055.
(24) Tedeschi RG, Calhoun LG: The Posttraumatic Growth
Inventory: Measuring the positive legacy of trauma. Journal
of Traumatic Stress. 1996, 9:455–472.
(25) Park CL, Cohen LH, Murch RL: Assessment and predic-
tion of stress-related growth. Journal of Personality. 1996,
(26) McMillen JC, Fisher RH: The Perceived Benefit Scales:
Measuring perceived positive life changes after negative
events. Social Work Research. 1998, 22:173–187.
(27) Tomich PL, Helgeson VS: Five years later: A cross-
sectional comparison of breast cancer survivors with
healthy women. Psycho-Oncology. 2002, 11:154–169.
(28) Antoni MH, Lehman JM, Kilbourn KM, et al.: Cognitive-
behavioral stress management intervention decreases the
prevalence of depression and enhances benefit finding
among women under treatment for early-stage breast
cancer. Health Psychology. 2001, 20:20–32.
(29) Park CL, Fenster JR: Stress-related growth: Predictors of
occurrence and correlates with psychological adjustment.
Journal of Social & Clinical Psychology. 2004, 23:195–215.
(30) Siegel K, Schrimshaw EW: Perceiving benefits in adversity:
Stress-related growth in women living with HIV=AIDS.
Social Science and Medicine. 2000, 51:1543–1554.
(31) Taylor SE, Lichtman RR, Wood JV: Attributions, beliefs
about control, and adjustment to breast cancer. Journal of
Personality and Social Psychology. 1984, 46:489–502.
(32) Carver CS, Antoni MH: Finding benefit in breast cancer
during the year after diagnosis predicts better adjustment
5 to 8 years after diagnosis. Health Psychology. 2004,
(33) Fromm K, Andrykowski MA, Hunt J: Positive and nega-
tive psychosocial sequelae of bone marrow transplantation:
Implications for quality of life assessment. Journal of
Behavioral Medicine. 1996, 19:221–240.
(34) Zoellner T, Maercker A: Posttraumatic growth in clinical
psychology – A critical review and introduction of a two
component model. Clinical Psychological Review. 2006,
(35) Taylor S: Adjustment to threatening events: A theory
of cognitive adaptation. American Psychologist. 1983,
(36) Schaefer JA, Moos RH: The context of posttraumatic
growth: Life crises, individual and social resources, and
coping. In Tedeschi RG, Park CL, Calhoun LG (eds),
Posttraumatic Growth: Positive Changes in the Aftermath
of Crisis. Mahwah, NJ: Lawrence Erlbaum Associates,
(37) Tedeschi RG, Calhoun LG: Posttraumatic growth: Concep-
tual foundations and empirical evidence. Psychological
Inquiry. 2004, 15:1–18.
(38) Ware JJ, Kosinski M, Gandek B: SF-36 Health Survey:
Manual and Interpretation Guide. Lincoln, RI: Quality-
Metrics Incorporated, 2000.
(39) Ware J, Kosinski M, Dewey J: How to Score Version 2 of
the SF-36 Health Survey. Lincoln, RI: QualityMetric Incor-
(40) Anderson JP, Kaplan RM, Coons SJ, Schneiderman LJ:
Comparison of the Quality of Well-being Scale and the
SF-36 results among two samples of ill adults: AIDS and
other illnesses. Journal of Clinical Epidemiology. 1998, 51:
(41) Baker C, Schuller DE: A functional status scale for measur-
ing quality of life outcomes in head and neck cancer
patients. Cancer Nursing. 1995, 18:452–457.
(42) Scheier MF, Carver CS, Bridges MW: Distinguishing opti-
mism from neuroticism (and trait anxiety, self-mastery, and
self-esteem): A re-evaluation of the Life Orientation Test.
Journal of Personality and Social Psychology. 1994,
(43) Blank TO, Bellizzi KM: After prostate cancer: Predictors
of well-being among long-term prostate cancer survivors.
Cancer. 2006, 106:2128–2135.
(44) Sherbourne C, Stewart A: The MOS social support survey.
Social Science Medicine. 1991, 32:705–714.
(45) Calhoun LG, Tedeschi RG: The foundations of posttrau-
matic growth: An expanded framework. In Calhoun LG,
Tedeschi RG (eds), Handbook of Posttraumatic Growth:
Research and Practice. Mahwah, NJ: Lawrence Erlbaum
Associates, 2006, 3–23.
(46) McBride CM, Ostroff JS: Teachable moments for promot-
ing smoking cessation: The context of cancer care and
survivorship. Cancer Control. 2003, 10:325–333.
(47) Petrella RJ, Koval JJ, Cunningham DA, Paterson DH: Can
primary care doctors prescribe exercise to improve fitness?
The step test exercise prescription (STEP) project. American
Journal of Preventive Medicine. 2003, 24:316–322.
198 Bellizzi et al.
Annals of Behavioral Medicine
(48) Mandelblatt J, Kanetsky PA: Effectiveness of interventions Download full-text
to enhance physician screening for breast cancer. Journal of
Family Practice. 1995, 40:162–171.
(49) Sallis JF, Hovell MF: Determinants of exercise behavior.
Exercise Sport Science Review. 1990, 18:301–330.
(50) Bellizzi KM: Expressions of generativity and posttraumatic
growth in adult cancer survivors. International Journal of
Aging and Human Development. 2004, 58:247–267.
(51) Blank TO, Bellizzi KM, Murphy K, Ryan K: How do men
‘‘make sense of’’ their prostate cancer?: Age and treatment
factors. Gerontologist. 2003, 43:342–343.
(52) Schulz U, Mohamed NE: Turning the tide: Benefit finding
after cancer surgery. Social Science and Medicine. 2004,
(53) Park CL, Helgeson VS: Introduction to the special section:
Growth following highly stressful life events—current status
and future directions. Journal of Consulting & Clinical
Psychology. 2006, 74:791–796.
(54) Calhoun LG, Tedeschi RG (eds): The Foundations of Post-
traumatic Growth: An Expanded Framework. Mahwah, NJ:
Lawrence Erlbaum Associates, 2006.
Volume 34, Number 2, 2007
Life Change and HRQOL199