Are Lifestyle Behavioral Factors Associated with Health-Related Quality of Life in Long-term Non-Hodgkin’s Lymphoma Survivors?

Abstract

Background

The objective of this study was to determine whether non-Hodgkin’s lymphoma survivors are meeting select American Cancer Society (ACS) health-related guidelines for cancer survivors, as well as to examine relationships between these lifestyle factors and health- related quality of life (HRQoL) and post-traumatic stress (PTS).

Methods

A cross-sectional sample of 566 NHL survivors was identified from the tumor registries of two large academic medical centers. Respondents were surveyed about physical activity, fruit and vegetable intake, body weight, tobacco use, HRQoL using the Medical Outcomes Study Short Form-36 and post-traumatic stress using the Post-traumatic Stress Disorder Checklist-Civilian form. Lifestyle cluster scores were generated based on whether individuals met health guidelines and multiple linear regression was used to evaluate relationships between lifestyle behaviors and HRQoL scores and PTS scores.

Results

11% of participants met all four ACS health recommendations. Meeting all four healthy recommendations was related to better physical and mental quality of life (β = 0.57, p <0.0001; β = 0.47, p = 0.002) and to lower PTS scores (β = −0.41, p = 0.01).

Conclusions

NHL survivors who met more ACS health-related guidelines appeared to have better HRQoL and less PTS. Unfortunately many survivors are not meeting these guidelines, which could impact their overall well-being and longevity.

Introduction

An estimated 558,000 individuals in the United States live with a history of non-Hodgkin’s lymphoma (NHL), a number projected to rise to over 710,000 by year 2022.^1–2 While improvements in treatments have led to longer survival, NHL survivors are at risk for physical and psychosocial adverse effects, including both long-term and late-effects that can negatively impact health-related quality of life (HRQoL). Long-terms effects include persistent fatigue, pain, depression, anxiety, and posttraumatic stress while late-effects include cardiac and pulmonary toxicities that manifest months or years after treatment ends.^3–6 Additionally, many cancer survivors are at higher than average risk for recurrence, secondary malignancies, and other co-morbid illnesses.^3,7 Further, life-style related factors including body weight, physical activity, and diet have an impact on risk for some cancers and other chronic diseases and may also affect HRQoL.

The American Cancer Society (ACS) has established guidelines for health behaviors for cancer survivors, which document evidence related to the benefits of engaging in health-related behaviors.^8–9 There is also a growing body of literature examining the prevalence of lifestyle-related cancer risk factors (e.g., overweight/obesity, poor dietary habits, inactivity, and smoking) among cancer survivors and their impact on HRQoL. Several studies have shown that compared to non-cancer controls, cancer survivors are no more likely to engage in recommended healthy lifestyle behaviors.^10–12 These studies have included mostly survivors of common cancers such as breast, prostate, colo-rectal, and cervical/uterine with little emphasis on NHL survivors.

Many studies have found that physical activity is related to improvements in HRQoL among cancer survivors, but only two focused on NHL survivors. Approximately 25% of NHL survivors have been found to meet current physical activity guidelines and they report higher HRQoL than those who are less active.^13–14 Overall, these individuals reported less fatigue, fewer symptoms from anemia and improved physical and functional well-being compared to those who were less active.^13–14

Considering that NHL survivors are at elevated risk for adverse long- and late-term side effects, recurrence, secondary cancers, and other chronic diseases, it is important to better understand their lifestyle behaviors and determine if appropriate health promotion interventions are indicated. Although prior studies have begun to investigate physical activity behaviors, there are limited data on other health-related factors in this population, most notably on diet, body weight, and tobacco use. Therefore, the aims of this study are to determine whether NHL survivors are meeting the ACS health-related guidelines for physical activity, fruit and vegetable intake, healthy weight, and tobacco use for cancer survivors, as well as to examine the relationships between lifestyle factors and HRQoL. We hypothesized that meeting more ACS health-related guidelines would be associated with better HRQoL.

Methods

Participants and Procedures

Follow-up surveys were mailed to 682 NHL survivors from two large academic centers who participated in a prior mailed survey as previously described.^15–16 A detailed comparison of participants and nonparticipants in the 2010 follow-up survey is described in the study by Smith et al.¹⁶ In brief, there were 566 participants in the follow-up survey and 144 nonresponders. Participants were more likely to be white, be older, married, have an income of more than $30,000, not have active disease, and to have reported better health-related quality of life in the initial survey. The Duke Cancer Center and University of North Carolina Lineberger Tumor Registries were used to identify potential participants following approval by the respective Institutional Review Boards; all participants provided written informed consent.

Measures

Demographic and clinical data were obtained via self-report; the 14-item Self-administered Comorbidity Questionnaire (SCQ) was also used.¹⁷ The SCQ has been used in hundreds of studies; its score, which ranges from 0–45, factors in the number of comorbidities as well as their severity and related functional impairment. However, only the number of comorbid conditions was used for this study. Of note, the SCQ is significantly associated with health status as measured by the Medical Outcomes Study Short Form-36 (SF-36).¹⁷

The survey included questions to evaluate whether participants met select ACS health behavior-related guidelines for physical activity, nutrition, healthy body mass index [BMI], and tobacco use. Because the survey was administered prior to the ACS’s 2012 survivorship guidelines on nutrition and physical activity, assessment of fruit and vegetable intake was based on servings/day (5-A-Day) versus current guidelines (i.e., cups a day).^8–9

Physical activity

Participants were asked about the number of times per week and minutes per time they exercised at a moderate-intensity level during a typical week.¹⁸ Responses were dichotomized as either meeting or not meeting physical activity guidelines.

Fruit and vegetable intake

Participants were asked: How many servings of vegetables do you usually eat or drink each day? Think of a serving as being about 1 cup of raw leafy vegetables, ½ cup of other cooked or raw vegetables, or ¼ cups of vegetable juice?; a similar question was asked to assess daily fruit intake.¹⁸ Responses were dichotomized (yes or no) based on whether participants met the 5-A-Day recommendation (i.e., consumed greater than or equal to five fruits and vegetables per day).

BMI

The calculation for BMI was derived from self-reported height and weight and categorized as follows: <18.5 (underweight), 18.5–24.9 (normal weight), 25–29.9 (overweight), or ≥30 (obese).

Tobacco use

Tobacco use was assessed through questions about smoking habits from the Center for Disease Control (CDC) Behavioral Risk Factor Surveillance System (BRFSS).¹⁹ The questions asked were: Have you smoked at least 100 cigarettes in your entire life?; Do you now smoke cigarettes every day, some days, or not at all?.

HRQoL (Health Status and Functioning)

To assess QOL, the Medical Outcomes Study Short Form-36 (SF-36; version 2.0) was used, which has become a common instrument for this construct¹⁶ even though it was initially developed to measure health status. The Medical Outcomes Study Short Form-36 contains 36 items grouped into eight subscales, four of which assess physical QOL (i.e., Physical component score [PCS] – physical functioning, role-physical, bodily pain, and general health) and four of which assess mental QOL (i.e., Mental component score [MCS] – vitality, social functioning, role-emotional, and mental health). The PCS and MCS were constructed based on U.S. population norms with the normed scores having a mean value of 50 and a standard deviation of 10.^20–21

Post-traumatic stress

The PTSD Checklist-Civilian form (PCL-C) is a 17-item scale that evaluates symptoms in the past month experienced in response to a trauma.²² Instructions were modified for the current study to key on symptoms in respect to diagnosis and treatment for lymphoma. The total score is the sum of all 17 items and ranges from 17–85.

Statistical Analysis

Descriptive statistics were used to examine participant’s demographic and clinical characteristics, lifestyle-related factors, HRQoL, and PTS. Lifestyle cluster scores were generated based on whether individuals met health guidelines for the following four lifestyle-related factors: ≥ 5 servings of fruits & vegetables/day, non-smoker, ≥ 150 minutes of moderate-intensity aerobic physical activity per week, and BMI < 25 kg/m². Each of these factors were dichotomized on whether a participant met the guideline or not (0 = no; 1 = yes) and were examined individually and by clusters (e.g., percentage of individuals meeting healthy guidelines for fruit & vegetable intake, physical activity, smoking, and BMI).

Bivariate analyses (chi-square analyses for categorical variables and analysis of variance for continuous variables) were conducted to examine relationships between demographics, clinical characteristics, SF-36 and PCL-C scores, and the number of health guidelines met for the lifestyle-related factors.

To address whether meeting a higher number of ACS health-related guidelines was associated with better HRQoL, multiple linear regression was used to investigate relationships between single lifestyle behaviors, lifestyle behavior cluster scores and SF-36 and PCL-C scores. Separate multiple linear regression analyses examined relationships between single lifestyle behaviors, lifestyle cluster scores and the outcome measures, SF-36 and PCL-C scores. The reference group for the single lifestyle-related variables was the group that did not meet the health guideline; similarly, for the lifestyle cluster scores the reference group was the lowest category (i.e., ≤ 1). Models were adjusted for demographic variables (i.e., age race, gender, marital status and education) and for current treatment status, reoccurrence in the last 5 years, and number of comorbidities. All analyses were performed using SAS version 9.3 (SAS Institute, Inc., Cary, NC).

Results

The mean age and number of years since diagnosis among participants was 67.2±12.5 years and 15.2±7.2, respectively (see Table 1). Most participants were Caucasian (87%), married or living with a partner (77%), retired (62%), and had less than a college education (57%). Clinically, 93% were not receiving active treatment, 75% did not have a recurrence within the past 5 years, and the mean number of comorbid conditions was 2.8. Most participants (67.5%) had two or more comorbid conditions. Concerning lifestyle-related factors, 64% were overweight or obese, 6% were current smokers, 56% were not consuming 5 or more fruits and vegetables per day and 52% did not meet the physical activity guidelines.

TABLE 1.

Participant characteristics (N=566)

Demographics	No. of respondents	%	Mean (SD)
Age			67.2 (12.5)
Gender
Male	272	48.1
Female	294	51.9
Race
Caucasian	494	87.3
African-American	51	9.0
Other	21	3.7
Ethnicity
Non-Hispanic	559	98.8
Hispanic	7	1.2
Marital status
Married or living with a partner	431	77.0
Not married	129	23.0
Employment status
Retired	341	62.4
Employed	180	33.0
Unemployed	25	4.6
Education
Less than college	316	56.6
College or post-graduate	242	43.4
Income level*
<$30,000	131	25.4
$30,000-$59,999	148	28.7
$60,000-$89,999	91	17.6
>$90,000	146	28.3
Clinical characteristics
Years since diagnosis			15.2 (7.2)
Current treatment status
Not in treatment	516	92.8
Receiving treatment	40	7.2
Recurrence in the past 5 years
No	413	75.2
One recurrence	58	10.6
More than one recurrence	49	8.9
Comorbid conditions
None	69	12.2
One	115	20.3
Two or more	382	67.5
Number of comorbidities			2.8 (2.1)
Lifestyle-related factors
Body mass index (BMI) – kg/m2			27.1 (5.7)
<18.5 (underweight)	9	1.6
18.5–24.9 (normal weight)	192	34.6
25–29.9 (overweight)	223	40.2
≥30 (obese)	131	23.6
Non-smoker	532	94.0
Intake of ≥ 5 fruits & vegetables/day	244	43.8
≥ 150 minutes of moderate-intensity physical activity/week	262	48.4
Quality of life factors
SF-36 PCS total score			45.0 (11.0)
SF-36 MCS total score			49.9 (10.9)
PTSD Checklist-Civilian form (PCL-C)			26.3 (9.6)

^*Missing data on 8.3% of participants.

Abbreviations: MCS, Mental Component Score; PCS, Physical Component Score; PTSD, Post-Traumatic Stress Disorder; SD, standard deviation; SF-36, Medical Outcomes Study Short Form-36.

With regard to lifestyle-related clusters, only 11% of participants met all four ACS health recommendations (i.e., non-smoker, normal BMI, ≥ 150 minutes of aerobic physical activity per week, and intake of ≥ 5 fruits and vegetables per day); 25% met three of the recommendations, 39% met a combination of two recommendations and 23% met only one recommendation, most of which were in the non-smoker category (see Figure 1). Table 2 shows the distribution of demographic factors, comorbidities, and HRQoL scores across the four ACS healthy lifestyle-related categories. For both genders, and also for Caucasians and African Americans, most respondents met two of the ACS guidelines. Significant differences existed between the number of comorbidities across the four guidelines (p = 0.019); participants meeting more healthy lifestyle recommendations had less comorbidities. There were no significant differences among those who had a recurrence in the last five years or based on years since diagnosis (data not included in Table 2). Significant differences existed in the mean SF-36 PCS and MCS scores among the four ACS guidelines (p = < 0.001) with higher scores corresponding with a greater number of ACS guidelines being met. Mean PCL-C scores were also significantly different (p = 0.003) revealing that lower scores (i.e., lower post-traumatic stress) were associated with meeting more healthy lifestyle-related guidelines. While SF-36 and PCL-C measure different dimensions of health, negative correlations exist between PCL-C scores and both MCS and PCS scores (i.e., −0.682 and −0.381, respectively).

Figure 1.

Health-related Behavior Guidelines Met

TABLE 2.

Characteristics of NHL Survivors by the Number of ACS Guidelines Met for Healthy Lifestyle-related Factors (N=566)^a

	Number of healthy lifestyle-related factors met				p-value
	≤ 1	2	3	4
Demographics	n %	n %	n %	n %
Gender
Male	65 (50.8)	109 (53.2)	60 (44.8)	19 (31.7)	0.011
Female	63 (49.2)	96 (46.8)	74 (55.2)	41 (68.3)
Race
Caucasian	105 (82.0)	179 (87.3)	124 (92.5)	55 (91.7)	0.019
African American	17 (13.3)	21 (10.2)	5 (3.7)	3 (5.0)
Other	6 (4.7)	5 (2.4)	5 (3.7)	2 (3.3)
Income
< $30,000	40 (34.2)	46 (24.2)	22 (18.3)	10 (17.4)	0.021
$30,000 – $59,999Ge	30 (25.6)	56 (29.5)	39 (32.5)	15 (25.9)	nde
$60,000 – $89,999	11 (9.4)	37 (19.5)	28 (23.3)	8 (13.8)
≥$90,000	36 (30.8)	51 (26.8)	31 (25.8)	25 (43.1)
Education
Less than college	87 (69.1)	108 (53.5)	70 (52.2)	25 (41.7)	0.0004
College or post-grad	39 (30.9)	94 (46.5)	64 (47.8)	35 (58.3)
Comorbidity problems
None	13 (10.2)	21 (10.2)	15 (11.2)	12 (20.0)	0.019
One	23 (18.0)	37 (18.1)	33 (24.6)	14 (23.3)
Two	92 (71.9)	147 (71.7)	86 (64.2)	34 (56.7)
	M (SD)	M (SD)	M (SD)	M (SD)	p-valueb
Age	66.4 (11.6)	66.6 (12.5)	68.1 (13.0)	67.6 (11.4)	0.631
SF-36 PCS T-Score	41.8 (11.1)	44.2 (11.1)	47.7 (10.2)	50.5 (8.4)	< 0.001
SF-36 MCS T-Score	46.4 (11.9)	50.5 (10.7)	51.9 (9.6)	52.7 (8.6)	< 0.001
PCL-C Total Score	28.5 (10.9)	25.9 (9.6)	25.0 (8.0)	23.7 (7.0)	0.003

Abbreviations: M, Mean; SD, standard deviation; SF-36 PCS, Medical Outcomes Study Short Form-36; PCS, Physical Component; PCL-C, Posttraumatic stress disorder checklist-Civilian form.

^aNumbers may not equal 566 due to missing data.

^bANOVA tested for significant differences in scores across categories.

Multivariable adjusted models (Tables 3–5) show associations between the healthy lifestyle-related clusters, as well as single lifestyle-related factors, and HRQoL. In the fully adjusted models, being a non-smoker (β = 0.37, p = 0.02) and exercising at least 150 minutes a week (β = 0.49, p <0.0001) were significant correlates of higher PCS scores. Also, meeting all four healthy recommendations and meeting three recommendations were related to better physical HRQoL (β = 0.57, p <0.0001; β = 0.49, p <0.0001, respectively). Non-smoker status, meeting physical activity and 5-A-Day recommendations were associated with better mental HRQoL (β = 0.71, p <0.0001 and β = 0.41, p <0.0001; β = 0.18, p = 0.03), as was meeting all four healthy recommendations (β = 0.47, p = 0.002). There was an inverse relationship with BMI < 25 kg/m², indicating that it was associated with lower mental HRQOL. Finally, meeting all four healthy lifestyle-related factors was related to lower post-traumatic stress scores (β = −0.41, p = 0.01) (see Table 5). Additionally, the single lifestyle behaviors non-smoker and meeting exercise guidelines (β = −0.73, p <0.0001 and β = −0.23, p = 0.01) were associated with lower scores.

TABLE 3.

Regression Coefficients for the Relationships between Healthy Lifestyle-related Factors and Quality of Life (SF-36 PCS)

SF-36 PCS (N=566)
Independent Variables	β (SE)	p-value (unadjusted)a	β (SE)	p-value (adjusted)b
Healthy Lifestyle-related Factors
No. of health recommendations met (clusters)
<1 (reference group)
2	0.16 (0.10)	0.12	0.15 (0.09)	0.10
3	0.52 (0.11)	<0.0001	0.49 (0.10)	<0.0001
4	0.77 (0.14)	<0.0001	0.57 (0.13)	<0.0001
Single health recommendations met
Smoking Status
Non-smoker	0.41 (0.17)	0.01	0.37 (0.15)	0.02
Exercise
≥ 150 minutes of mod act/week	0.59 (0.08)	<0.0001	0.49 (0.07)	<0.0001
BMI
Body mass index: < 25kg/m2	0.14 (0.09)	0.09	0.08 (0.08)	0.29
Fruit and Vegetable consumption
≥ 5 fruits and vegetables/day	0.15 (0.08)	0.07	0.12 (0.07)	0.12

Abbreviations: SF-PCS, Medical Outcomes Study Short Form-36; PCS, Physical Component; β, standardized regression coefficient; SE, standard error; mod act/week, moderate activity per week.

^aAdjusted for DEMO (demographic) variables age, race, gender, marital status, education.

^bAdjusted for DEMO variables and CLIN (clinical characteristics) variables current treatment status, reoccurrence in last 5 years, comorbidities. Healthy lifestyle-related factors compared to the reference group (i.e., smoker; physical activity: < 150 minutes of moderate intensity activity/week; BMI: ≥ 25kg/m²; fruit and vegetables/day: < 5 fruits and vegetables/day).

TABLE 5.

Regression Coefficients for the Relationships between Healthy Lifestyle-related Factors and Posttraumatic Stress Disorder Scores (PCL-C)

PCL-C (N=566)
Independent Variables	β (SE)	p-value (unadjusted)a	β (SE)	p-value (adjusted)b
Healthy Lifestyle-related Factors
No. of health recommendations met (clusters)
<1 (reference group)
2	−0.26 (0.11)	0.02	−0.28 (0.10)	0.01
3	−0.32 (0.12)	0.01	−0.30 (0.11)	0.01
4	−0.50 (0.15)	0.001	−0.41 (0.15)	0.01
Single health recommendations met
Smoking Status
Non-smoker	−0.85 (0.18)	<0.0001	−0.73 (0.17)	<0.0001
Exercise
≥ 150 minutes of mod act/week	−0.30 (0.09)	0.001	−0.23 (0.08)	0.01
BMI
Body mass index: < 25kg/m2	0.08 (0.09)	0.38	0. 06 (0.09)	0.47
Fruit and Vegetable consumption
≥ 5 fruits and vegetables /day	−0.11 (0.09)	0.18	−0.08 (0.08)	0.35

Abbreviations: PCL-C, Posttraumatic Stress Disorder Checklist-Civilian form; β, standardized regression coefficient; SE, standard error; mod act/week, moderate activity per week.

^aAdjusted for DEMO (demographic) variables age, race, gender, marital status, and education.

^bAdjusted for DEMO variables and CLIN (clinical characteristics) variables current treatment status, reoccurrence in last 5 years, comorbidities. Healthy lifestyle-related factors compared to the reference group (i.e., smoker; physical activity: < 150 minutes of moderate activity/week; BMI: ≥ 25kg/m²; fruit and vegetables/day: < 5 fruits and vegetables/day).

Discussion

This study examined the prevalence of meeting the ACS health behavior-related recommendations and associations with HRQoL in NHL survivors. Findings indicate that while most were non-smokers (94%), very few survivors met all four of the health recommendations. In fact, only 36% were meeting recommendations for normal body weight, 44% for 5-A-Day and 48% for physical activity. However, data suggest that participants who met three or four ACS recommendations (met by 25% and 11%, respectively) had better HRQoL and less stress than those meeting fewer recommendations. In terms of individual behaviors, the greatest association with HRQoL related to non-smoking status and exercising ≥ 150 minutes a week compared with BMI and fruit and vegetable consumption.

While our sample appears to have similar health behaviors to the general population (with the exception of smoking), the fact that they are not meeting the ACS guidelines is concerning since many survivors are at higher than average risk for co-morbid conditions, such as cardiovascular disease, and secondary cancers. These late effects could potentially be mitigated through healthy lifestyle practices.²³ In fact, Jones et al.²⁴ found that Hodgkin lymphoma survivors who met national guidelines for vigorous exercise (i.e., 3 sessions/week of ≥ 20 minutes in duration) were less likely to have a cardiovascular event compared to those not meeting the guideline. In our sample, the range for meeting national physical activity guidelines was between 30–46%, compared to 52% in the general population.^11,25 A lower percentage of our sample smoke compared to the general population (6% and 21%, respectively).^10–12 Weaver et al.²³ found that long-term survivors of breast, prostate, colorectal, and gynecologic cancers reported 1.6 cardio-vascular risk factors (e.g., smoking, unhealthy BMI, physical inactivity). They also found little differences in risk factors between the cancer types, however differences existed based on age and race/ethnicity (i.e., African-Americans and Hispanics were more likely to report higher prevalence of risk factors).²³ We found that Caucasians were more likely to meet all four of the ACS health recommendations but found no difference by age. This may be due to the fact that we had an older sample. Our finding that Caucasians were more likely than African Americans to meet a greater number of health guidelines is consistent with those reported by Findley and colleagues.²⁶ However, there was low representation of African American survivors in both studies. We also found that females, as well as participants with higher education and income were more likely to engage in a greater number of health behaviors. Among diverse groups of cancer survivors, predictors of healthier lifestyle practices were female gender and higher education.^11,26 Cancer disparities exist among minorities and those with lower socioeconomic status and unhealthy lifestyle behaviors have been identified to be contributing factors.²⁷ Further research is needed to develop effective targeted behavioral interventions to improve lifestyle behaviors among African Americans survivors and those with low socioeconomic status.

Our PCS and MCS mean scores were comparable to those in the study by Bellizzi and colleagues.¹³ Also, we found that specific behaviors, such as exercise and non-smoking, were associated with higher physical HRQoL scores among NHL survivors. Findings from the literature support that meeting current exercise guidelines has a positive impact on NHL survivors’ quality of life.^13–14 With regard to smoking status, female cancer survivors who were non-smokers have reported better physical and mental HRQoL compared to those who were current smokers.²⁸ Of additional interest, meeting physical activity guidelines and non-smoking status corresponded to lower posttraumatic stress (i.e., better HRQoL) in our sample. These are important findings that provide further support for research focused on behavior change in the area of exercise and smoking among NHL survivors.

Our findings that lower PCL-C scores were associated with healthier behaviors (e.g., non-smoking) is consistent with findings among other populations including adult survivors of childhood cancers.²⁹ Although there is a paucity of evidence-based strategies to reduce cancer-related PTSD, cognitive behavioral therapy, which often includes an emphasis on healthy behaviors such as exercise, nutrition and avoidance of substance abuse, is beneficial in improving PTSD symptoms in the general population.³⁰

This study especially adds to the literature in finding that HRQoL increases as the number of health guidelines met increases, even though the individual factors of meeting the 5-A-Day and body weight recommendations alone did not significantly relate to higher HRQoL. It is possible that individuals who met more health-related guidelines were displaying additional health-related behaviors that were not assessed in this study. That said, typically a healthy diet and normal body weight reduce an individual’s risk for co-morbid illnesses such as cardiovascular disease, and many NHL survivors are already at higher than average risk for cardiovascular disease as a result of cardio-toxic treatments. Additionally, many behavioral risk factors increase an individual’s risk for secondary cancers. Therefore, it is prudent to include these healthy lifestyle-related factors in research and educational programs addressing behavior change in this population, especially among those who received treatments known to affect the cardiovascular system.

Strengths of this study include a large sample of NHL survivors with a mean HRQoL score comparable to another large population-based sample of NHL survivors (i.e., enhances generalizability of findings).¹³ In addition, the high response rate (83%) reduces the chance for non-responder bias. Also, this was the first report to investigate multiple lifestyle-related factors as they relate to HRQoL among long-term NHL survivors. Study limitations include the inability to draw inferences about causality between HRQoL and health behaviors due to the cross-sectional design. Also, the use of self-report for health behaviors may have led to reporting bias (e.g., inflated estimates of the prevalence of healthy behaviors). While it was beyond the scope of this current study to analyze NHL survivors by subtype, we recognize that this may be a limitation of the findings considering the many different subtypes are known to have very different clinical characteristics and require different types of treatment approaches that affect long-term health outcomes. However, our findings did reveal that when clinical characteristics (i.e., current treatment status, recurrence in past 5 years, and comorbidity scores) were controlled for in our adjusted models the results were essentially the same as those in the unadjusted models indicating that clinical burden of disease did not appear to effect whether individuals met health-related guidelines. Future studies addressing the relationship between lifestyle behaviors and HRQoL among NHL survivors that stratify analyses based on at least the most common subtypes of NHL may be warranted.

In conclusion, this study indicated that healthy lifestyle behaviors related to better HRQoL among long-term NHL survivors. However, the majority of survivors are not meeting ACS guidelines for health behaviors, which could impact their overall well-being and longevity. Additional research is warranted in the area of health behavior change, which could lead to improvements in educational efforts to enhance the lives of long-term NHL survivors.

TABLE 4.

Regression Coefficients for the Relationships between Healthy Lifestyle-related Factors and Quality of Life (SF-36 MCS)

SF-36 MCS (N=566)
Independent Variables	β (SE)	p-value (unadjusted)a	β (SE)	p-value (adjusted)b
Healthy Lifestyle-related Factors
No. of health recommendations met (clusters)
<1 (reference group)
2	0.36 (0.11)	0.001	0.36 (0.11)	0.001
3	0.44 (0.12)	0.0003	0.41 (0.12)	0.001
4	0.59 (0.16)	0.0002	0.47 (0.15)	0.002
Single health recommendations met
Smoking Status
Non-smoker	0.75 (0.18)	<0.0001	0.71 (0.18)	<0.0001
Exercise
≥ 150 minutes of mod act/week	0.47 (0.09)	<0.0001	0.41 (0.08)	< 0.0001
BMI
Body mass index: < 25kg/m2	−0.17 (0.09)	0.06	−0.21(0.09)	0.02
Fruit and Vegetable consumption
≥ 5 fruits and vegetables/day	0.22 (0.09)	0.01	0.18 (0.09)	0.03

Abbreviations: SF-MCS, Medical Outcomes Study Short Form-36; MCS, Mental Component; β, standardized regression coefficient; SE, standard error; mod act/week, moderate activity per week.

^aAdjusted for DEMO (demographic) variables age, race, gender, marital status, and education.

^bAdjusted for DEMO variables and CLIN (clinical characteristics) variables current treatment status, reoccurrence in last 5 years, comorbidities. Healthy lifestyle-related factors compared to the reference group (i.e., smoker; physical activity: < 150 minutes of moderate activity/week; BMI: ≥ 25kg/m²; fruit and vegetables/day: < 5 fruits and vegetables/day).