## Abstract

Background: We aimed to assess anxiety and the psychological impact of routine surveillance scans in long-term survivors of adult aggressive lymphoma.

Patients and methods: In this cross-sectional observational study of 70 survivors of curable adult aggressive lymphoma, we measured anxiety and the doctor–patient relationship and performed a qualitative interview (n = 30) focused on patient perception of routine follow-up imaging studies.

Results: Participants were diagnosed with aggressive lymphoma a median of 4.9 years (2.4–38.0 years) before enrollment. Thirty-seven percent of patients were found to meet criteria for clinically significant anxiety, which was not associated with years since diagnosis. In multivariate analysis, history of relapse and a worse doctor–patient relationship were independently associated with higher anxiety levels. Despite representing a largely cured population, in qualitative interviews patients reported fear of recurrence as a major concern and considerable anxiety around the time of a follow-up imaging scan.

Conclusions: Routine surveillance scans exacerbate underlying anxiety symptoms and fear of recurrence in survivors of aggressive lymphoma. Strategies to minimize follow-up imaging and to improve doctor–patient communication should be prospectively evaluated to address these clinically significant issues.

## introduction

Surveillance imaging scans in asymptomatic survivors of aggressive lymphoma are generally accepted as standard of care as relapsed disease is potentially curable. However, the value of routine testing is questionable because most relapses are detected by the patient or physician without the use of a scan [1–4]. Imaging can lead to false-positive findings, requiring further testing and costs [5]. Additional concerns have been raised regarding radiation exposure associated with computed tomography (CT) scans, particularly the risk of secondary cancers [6, 7]. Lastly, surveillance scans do not appear to be cost-effective in Hodgkin’s lymphoma (HL) survivors [8].

In addition, there may be a psychological burden. After cancer treatment, patients are left with many uncertainties, including fear of recurrence, treatment effects, and others [9]. These anxieties can be heightened around the time of a scan. The true psychological impact of surveillance scans in asymptomatic cancer survivors has not been previously evaluated.

The doctor–patient relationship is at the crux of cancer care. Patient-centered care, as defined by the National Cancer Institute (NCI), involves mutual trust, respect, and commitment and is facilitated by communication [10]. The goal of high-quality communication is to enhance patient well-being. However, there are few studies that measure the impact of patient-centered communication on health outcomes such as anxiety, particularly in long-term survivors.

Patients with aggressive lymphoma have a high cure rate, generally undergo numerous scans after treatment, and therefore are an ideal population to study. We hypothesized that anxiety in lymphoma survivors was related to fear of recurrence and that surveillance scans were a source of anxiety. Furthermore, better patient-directed communication would alleviate some of these fears.

## methods

### study design and sample

The Institutional Review Board (IRB) of Weill Cornell Medical Center (WCMC) granted approval of the study. Recruitment took place in the outpatient practice by physicians or nurses, based on patient records and databases. Patients external to WCMC contacted the principal investigator after learning of the study on www.clinicaltrials.gov, and therefore, an IRB modification was granted for participation of non-WCMC patients. The data presented are from a parent study of quality of life in lymphoma survivors.

Eligible participants were survivors of HL and aggressive non-Hodgkin’s lymphoma (NHL) subtypes, including diffuse large B-cell lymphoma (DLBCL), follicular grade 3, Burkitt’s, etc. Inclusion criteria included last lymphoma treatment >2 years, age ≥18 at diagnosis, and no known cancer. Exclusion criteria included indolent (noncurable) lymphoma and non-English speaking. All participants provided informed consent.

### survey content

Patients were interviewed by the primary investigator, and assessed for demographic, socioeconomic, and lymphoma-specific data. Medical records of the WCMC patients were abstracted for verification. Other survey data included the following.

#### patient–physician relationship scale.

Items were drawn from on the NCI monograph on patient-centered communication in cancer care [10]. Patients were asked to confidentially evaluate their hematologist/oncologist and rank their agreement with the following statements: “I feel understood by my physician,” “I actively participate in decision-making with my physician,” “My physician helps me cope with uncertainty,” “I trust my physician,” “I feel comfortable talking to my physician about my concerns,” and “My physician provides me with information that helps me to better understand my diagnosis and treatment options.”

#### state-trait anxiety index.

The State-Trait Anxiety Index (STAI) was used to measure anxiety symptoms [11]. This is a two-part index that measures state (transitory feelings of apprehension) and trait (a person’s tendency to perceive situations as threatening) anxiety, each of which has 20 items. The range of scores is 20–80, with a higher score indicating higher anxiety. This index has been shown to be reliable [11] and valid [12]. Consistent with previous literature, we used a cut-off of 40 on either subscale to define clinically significant anxiety [13].

#### data analysis.

Descriptive statistics were computed for demographic and lymphoma-specific data. The STAI [11] was scored according to the appropriate algorithm. For the patient–physician relationship questionnaire, “definitely true” was scored as 0, “mostly true” as 1, “don't know” as 2, “mostly false” as 3, and “definitely false” as 4. The responses were added together, with a lower score indicating a better relationship.

Statistical comparisons were made using the Student’s t-test. Linear regression was used to determine associations with anxiety. Quantitative analysis was done in STATA™ 8.2 statistical software (StataCorp LP, College Station, TX) and P < 0.05 were considered statistically significant.

#### qualitative interview.

Qualitative methodology is appropriate where little is known about the issue a priori. We included a series of qualitative questions to learn about the experience of lymphoma survivorship, including the impact of follow-up scans. The responses to the questions, “Tell me what it is like for you to have a follow-up CT scan?” “How does it affect you emotionally?” and “What downsides are there for you?” are presented here.

The first 30 qualitative interviews were recorded and transcribed. These were analyzed via grounded theory by one of the two investigators. Each transcript was reviewed by the other investigator, and if discrepancy occurred, a consensus was reached. A third investigator reviewed all results.

## results

### demographics

Between April and November 2008, 80 eligible patients were referred to the study. Of these, 70 patients completed the interview, 57 were from WCMC and 13 were external. Of the seven patients from WCMC who declined, five “did not have time” and two were “too anxious.” These seven patients were no different from those who completed the interview on the basis of gender (P = 0.07), diagnosis (P = 0.83), and years since diagnosis (P = 0.72). However, those who declined were older than those who enrolled (mean 62 versus 48 years, P = 0.04). There were three external patients who did not respond after initial contact and whose demographic information was not available for comparison. The participating WCMC patients and external patients did not differ according to age (P = 0.75), gender (P = 0.09), and years since diagnosis (P = 0.64).

Patients were a median 47 years of age (range 23–80) and were diagnosed with lymphoma a median of 4.9 years (2.4–38.0) before enrollment. Other demographic information is described in Table 1.

Table 1.

Demographics

 n (%) Age (years) ≤20 0 (0) 20–29 8 (11) 30–39 18 (26) 40–49 12 (17) 50–59 16 (23) 60–69 6 (9) 70–79 9 (13) ≥80 1 (1) Gender Male 25 (36) Female 45 (64) Race Caucasian 57 (83) African American 4 (6) Hispanic/Latino 7 (10) Other 1 (1) Yearly income <$25K 9 (13)$25–50K 12 (17) $50–75K 8 (11)$75–100K 11(18) >$100K 30 (43) Education completed <12th grade 2 (3) High school 13 (19) College 37 (53) Graduate school 18 (26) Current relationship status Single 22 (31) Married 37 (53) Divorced 7 (10) Widowed 4 (6) Have children 38 (54)  n (%) Age (years) ≤20 0 (0) 20–29 8 (11) 30–39 18 (26) 40–49 12 (17) 50–59 16 (23) 60–69 6 (9) 70–79 9 (13) ≥80 1 (1) Gender Male 25 (36) Female 45 (64) Race Caucasian 57 (83) African American 4 (6) Hispanic/Latino 7 (10) Other 1 (1) Yearly income <$25K 9 (13) $25–50K 12 (17)$50–75K 8 (11) $75–100K 11(18) >$100K 30 (43) Education completed <12th grade 2 (3) High school 13 (19) College 37 (53) Graduate school 18 (26) Current relationship status Single 22 (31) Married 37 (53) Divorced 7 (10) Widowed 4 (6) Have children 38 (54)

### diagnosis

The most common histology in this group was HL (49%), followed by DLBCL (39%), follicular grade 3 lymphoma (6%), and other (7%).

### treatment history

Patients had received a median of 6.0 (range 1–15) chemotherapy cycles, 43% received radiation, 9% suffered a relapse (now in remission), and 6% underwent transplant.

### patient–physician relationship

The mean score was 3.59 (SD 3.66) on a scale of 0–24, with a lower number indicating a better relationship. Eighty-six percent of participants stated that they feel understood by their physician (response “definitely true” or “mostly true”), 81% actively participate in decision making with their physician, 74% feel their physician helps them cope with uncertainty, 96% trust their physician, 93% feel comfortable talking to their physician about their concerns, and 93% stated their physician provides information that helps them to better understand their diagnosis and treatment options.

### anxiety

The mean STAI score was 65.8 (±21.0), while the mean of the State subscale was 30.4 (±10.9) and the mean of the Trait subscale was 35.4 (±11.3), on a scale of 20–80, with a higher score indicating more anxiety. In total, 26 participants (37%) had clinically significant anxiety symptoms.

### associations with anxiety

Univariate linear regression analysis was performed to find associations between demographic, socioeconomic, disease factors, and anxiety. Higher levels of anxiety were seen with a history of relapse and a worse doctor–patient relationship (P < 0.05). There was a trend for women to have a higher mean anxiety score (mean 69.3, SD 20.4) than men (mean 59.6, SD 21.0; P = 0.06). Age, employment status, having children, income, marital status, race, education, radiation treatment, number of treatment regimens, histological diagnosis, and years since diagnosis were not associated with anxiety. Results are presented in Table 2.

Table 2.

Univariate associations with anxiety, in order of magnitude. Regression coefficients and standard errors are listed.

 Correlate Coefficient (standard error) P value Demographics Gender −9.67 (5.14) 0.06 Age −0.26 (0.15) 0.10 Employment status 9.10 (5.01) 0.07 Having children −3.22 (2.08) 0.13 Income −1.69 (1.69) 0.32 Marital status −2.21 (2.89) 0.45 Race −1.82 (2.48) 0.47 Education level 1.70 (3.38) 0.62 Treatment data Relapse 24.08 (8.54) 0.006 Radiation treatment 6.10 (5.05) 0.23 No. treatment regimens 2.74 (2.58) 0.29 Diagnosis −2.21 (2.91) 0.45 Years since diagnosis −0.22 (0.46) 0.63 Months since last scan 0.32 (0.10) 0.75 Doctor–patient relationship 2.31 (0.64) 0.001
 Correlate Coefficient (standard error) P value Demographics Gender −9.67 (5.14) 0.06 Age −0.26 (0.15) 0.10 Employment status 9.10 (5.01) 0.07 Having children −3.22 (2.08) 0.13 Income −1.69 (1.69) 0.32 Marital status −2.21 (2.89) 0.45 Race −1.82 (2.48) 0.47 Education level 1.70 (3.38) 0.62 Treatment data Relapse 24.08 (8.54) 0.006 Radiation treatment 6.10 (5.05) 0.23 No. treatment regimens 2.74 (2.58) 0.29 Diagnosis −2.21 (2.91) 0.45 Years since diagnosis −0.22 (0.46) 0.63 Months since last scan 0.32 (0.10) 0.75 Doctor–patient relationship 2.31 (0.64) 0.001

### multivariate analysis

In multivariate regression analysis, the doctor–patient relationship (P = 0.02) and history of relapse (P = 0.02) remain independently associated with anxiety.

### follow-up scans

Last follow-up scan was a mean of 14.8 months (SD 25.3) before the interview. Patients were asked their opinion about the number of scans to monitor for recurrence, and 84% of participants indicated that the number was “just right,” 4% felt that there were “not enough,” and 12% indicated that there were “too many.”

### qualitative interviews

Almost all of the lymphoma survivors surveyed experience the fear of recurrence and reported that this fear became heightened around the time of follow-up scans. Although a few participants said that the scans “don't phase me,” for the majority, the scans cause significant anxiety. One patient referred to the experience as “scan-itis.” One participant stated that they are “terrorizing” and while not all used such strong words, most felt that the scans are “a big thing.” The time between having the scan and receiving results is particularly difficult and was described as “whole other kind of fear.” Others disliked the inconvenience and the intravenous contrast. Concern about radiation exposure was common, and this led some to feel they were overtested.

However, there was a positive side as well—reassurance that the disease is in remission. Another stated, “I have a lease on life for one year and I can start all over again.” Therefore, scans are a double-edged sword. “They are fine ’cause they tell the doctor if there's anything wrong with me, but as far as the experience I really hate it.”

Anxiety for most patients was reported to peak before a scan and drop following the doctor visit. This was explained by one, “Even if you know that the statistics are in your favor, and you know that you're feeling all right, you still have a huge amount of anxiety that goes away, oddly enough, it goes away immediately when you hear that it's fine. It's a complete relief.” The scans can be a reminder of the cancer. “The week leading up to it is very nerve-racking, and it's a lot of the thoughts that I had when I was first diagnosed start to come back…once it's over, a sense of relief.”

## discussion

In this study of long-term survivors of aggressive lymphoma, we found a 37% incidence of clinically significant anxiety symptoms and a fear of recurrence that became especially prevalent in the time leading up to surveillance scans, based on qualitative interviews. The only factors independently associated with anxiety symptoms in multivariate analysis were the doctor–patient relationship, where a better relationship correlated with less anxiety symptoms, and relapse, where a history of relapse was associated with greater anxiety symptoms.

Of the participants, 37% reported clinically significant anxiety symptoms, double that of the general population [14]. Anxiety has not been previously studied extensively in lymphoma survivors. The largest study of 557 HL survivors found a 14.5% incidence of anxiety using the Hospital Anxiety and Depression Scale, higher than the general Norwegian population [15]. Other studies have measured psychological distress using the Brief Symptom Inventory, with 21% of HL survivors reporting a high score [16]. In our study, interviews tended to be around the time of a follow-up visit, and therefore, we recognize that it is possible that anxiety level is higher than if the interviews occurred at a different time. Alternatively, our survey may have found higher levels of anxiety than other studies due to a different population or screening instrument, or using a more conservative cut-off for clinically significant symptoms. Nonetheless, even transient significant anxiety levels are important and warrant consideration of potential interventions.

There were no demographic parameters that were correlated with anxiety in multivariate analysis. Survivors who had a history of relapsed lymphoma had higher anxiety levels than others, suggesting this is a population at risk. In this cross-sectional study, anxiety symptoms did not decrease with increasing years since diagnosis, indicating that these issues are not limited to the first few years after treatment. A longitudinal study to explore these issues is warranted.

Anxiety specifically regarding surveillance scans in cancer survivors has not previously been well characterized in the literature. A study of survivors of breast, colorectal, and prostate cancer found that 36%–44% express worries about testing, and cancer-related health worries (including worry about future diagnostic tests) were the strongest predictor of anxiety and depression [9]. To our knowledge, there are no data on anxiety and diagnostic testing in lymphoma survivors. This study has value in elucidating the focus of anxiety symptoms in survivors of aggressive lymphoma, which are linked to fear of recurrence around the time of a routine surveillance scan based on our data. Many survivors reported that anxiety symptoms worsen before a follow-up appointment, peak leading up to obtaining the results of the scan, and lessen immediately afterwards. It is hypothesized that many lymphoma survivors have underlying anxiety symptoms, which are exacerbated by surveillance scans.

Surveillance scans in lymphoma survivors are performed frequently, despite no randomized evidence that they lead to improved outcomes. Scans can lead to false-positive findings and increased testing, high cost to our health care system, and radiation exposure. In addition, we have demonstrated another important adverse effect—scans are a significant source of anxiety for patients. We believe this provides further evidence of the potential value of minimizing surveillance scans in the asymptomatic lymphoma survivor.

We show that good patient-centered communication lessens anxiety. Previous studies of physician–patient communication have shown the ability to improve patient health outcomes, including anxiety [17], although most studies measure a single communication episode and we measured the patient’s perception of the overall relationship. Physicians should recognize that fear of relapse is the greatest worry in this group of lymphoma survivors despite the actual risk being very low, and this fear was irrespective of the number of years since treatment. While in general, patients had a very favorable perception of their physician, only 74% felt that their physician helps them cope with uncertainty. Our data support a strategy to help survivors manage fears and provide education and reassurance, particularly around the time of follow-up scans. In particular, reiteration of the true risk of relapse to patients may be helpful in relieving fears of patients who harbor an inaccurately high perception of risk. Utilizing a survivorship care plan as a tool to deliver this message and minimize anxiety may be one useful strategy.

Strengths of this study include both the qualitative and quantitative methods used, which provided insight into patients’ views on surveillance scans, an understudied area. Limitations include the cross-sectional design, so cause and effect conclusions cannot be drawn from the associations. It is a small study with the majority of patients from one academic medical center. Lastly, the measurement of the doctor–patient relationship is novel and therefore the validity and reliability are unknown.

In conclusion, we believe that our data provide compelling rationale for prospective studies of less frequent follow-up imaging in patients with aggressive lymphoma after therapy, and question whether surveillance scans in asymptomatic patients with no clinical evidence of recurrence offer any benefit. It is possible that the harm of routine surveillance scans for survivors of aggressive lymphoma may outweigh the value, given the lack of randomized data on the effectiveness of the current practice standards, false-positive findings, high cost, radiation exposure, and negative emotional impact on patients.

## funding

Agency for Healthcare Research and Quality (5 T32 HS000066) and the NCI at the National Institutes of Health (K12CA 90628-8).

The authors would like to thank Laura Robbins, DSW, ACSW, MSW for her contributions to the qualitative portion of this study.

## References

1.
Edelman
MM
Meyers
FJM
Siegel
D
, et al.  .
The utility of follow-up testing after curative cancer therapy. A critical review and economic analysis
J Gen Intern Med
,
1997
, vol.
12
(pg.
318
-
331
)
2.
Dryver
ET
Jernstrom
H
Tompkins
K
, et al.  .
Follow-up of patients with Hodgkin’s disease following curative treatment: the routine CT scan is of little value
Br J Cancer
,
2003
, vol.
89
(pg.
482
-
486
)
3.
Xavier
MF
Schuster
SJ
C
, et al.  .
Detection of relapse in diffuse large B-cell lymphoma (DLBCL) and Hodgkin’s lymphoma (HL): observations and implications for post-remission radiologic surveillance
Blood (ASH Annual Meeting Abstracts)
,
2006
, vol.
108

2428
4.
Foltz
LM
Song
KW
Connors
JM
Who actually detects relapse in Hodgkin lymphoma: patient or physician?
Blood (ASH Annual Meeting Abstracts)
,
2004
, vol.
104

3124
5.
Seam
P
Juweid
ME
Cheson
BD
The role of FDG-PET scans in patients with lymphoma
Blood
,
2007
, vol.
110
(pg.
3507
-
3516
)
6.
Brenner
DJ
Hall
EJ
Computed tomography—an increasing source of radiation exposure
N Engl J Med
,
2007
, vol.
357
(pg.
2277
-
2284
)
7.
Fazel
R
Krumholz
HM
Wang
Y
, et al.  .
Exposure to low-dose ionizing radiation from medical imaging procedures
N Engl J Med
,
2009
, vol.
361
(pg.
849
-
857
)
8.
BA
Punglia
RS
Kuntz
KM
, et al.  .
Cost-effectiveness analysis of computerized tomography in the routine follow-up of patients after primary treatment for Hodgkin's disease
J Clin Oncol
,
2006
, vol.
24
(pg.
4116
-
4122
)
9.
Deimling
GT
Bowman
KF
Sterns
S
, et al.  .
Cancer-related health worries and psychological distress among older adult, long-term cancer survivors
Psychooncology
,
2006
, vol.
15
(pg.
306
-
320
)
10.
Epstein
R
Street
RJ
Patient-Centered Communication in Cancer Care. Promoting Healing and Reducing Suffering

Bethesda, MD: National Cancer Institute, 2007; NIH Publication No. 07–6225
11.
Spielberger
C
Manual for the State-Trait Anxiety Inventory (STAI)
,
1983
Palo Alto, CA
Consulting Psycologists Press
12.
Spielberger
C
Reheiser
E
Ritterband
L
, et al.  .
Assessment of emotional states and ersonality traits: measuring psychological vital signs
Clinical Personality Assessment: Practical Approaches. Edition
,
1995
New York
Oxford University Press
13.
Teixeira
JMA
Fisk
NM
Glover
V
Association between maternal anxiety in pregnancy and increased uterine artery resistance index: cohort based study
BMJ
,
1999
, vol.
318
(pg.
153
-
157
)
14.
Kessler
RC
Chiu
WT
Demler
O
Walters
EE
Prevalence, severity, and comorbidity of 12-month DSM-IV disorders in the National Comorbidity survey replication 10.1001/archpsyc.62.6.617
Arch Gen Psychiatry
,
2005
, vol.
62
(pg.
617
-
627
)
15.
Loge
JH
Abrahamsen
AF
Ekeberg
O
, et al.  .
Psychological distress after cancer cure: a survey of 459 Hodgkin's disease survivors
Br J Cancer
,
1997
, vol.
76
(pg.
791
-
796
)
16.
Kornblith
AB
Herndon
JE
2nd
Zuckerman
E
, et al.  .
Comparison of psychosocial adaptation of advanced stage Hodgkin's disease and acute leukemia survivors. Cancer and Leukemia Group B
Ann Oncol
,
1998
, vol.
9
(pg.
297
-
306
)
17.
Stewart
MA
Effective physician-patient communication and health outcomes: a review
CMAJ
,
1995
, vol.
152
(pg.
1423
-
1433
)