Living with chronic headache
Ciere, Yvette
IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from it. Please check the document version below.
Document Version
Publisher's PDF, also known as Version of record
Publication date: 2018
Link to publication in University of Groningen/UMCG research database
Citation for published version (APA):
Ciere, Y. (2018). Living with chronic headache: A personal goal and self-regulation perspective. University of Groningen.
Copyright
Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).
Take-down policy
If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.
Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the number of authors shown on this cover page is limited to 10 maximum.
Distinct trajectories of positive and negative
affect after colorectal cancer diagnosis
Yvette Ciere, Moniek Janse, Josué Almansa, Annemieke
Visser, Robbert Sanderman, Mirjam Sprangers, Adelita
Ranchor, Joke Fleer
Health Psychology, 2017, 36(6)
Chapter
Objective:Insight into trajectories of positive and negative affect across the cancer continuum may improve understanding of the nature of adjustment problems. The primary aim of this study was to identify subgroups of patients with distinct trajectories of positive and negative affect following diagnosis of colorectal cancer. Secondary to this aim, the co-occurrence between trajectories and their association with goal-related processes was explored.
Methods: Colorectal cancer patients (n=186) completed questionnaires within one month,
7 months, and 18 months after diagnosis. Multilevel models were used to study the trajectory of positive and negative affect, as measured with the PANAS.
Results: Four classes with distinct positive affect trajectories were identified: low (18.8%),
increasing (6.7%), moderate (68.2%) and high (6.3%); two trajectories of negative affect
emerged: low (36.3%) and moderate (63.7%). There was no significant association between positive and negative trajectory class probabilities. The average trajectory of positive affect co-varied with levels of goal disturbance and goal re-engagement over time, while the average negative affect trajectory co-varied with goal disturbance and goal disengagement.
Conclusions: Compared to the general population, our sample of cancer patients suffered
from a lack of positive emotions, but not a high presence of negative emotions. About one-fifth of patients reported low positive affect up to 18 months after diagnosis and may benefit from supportive care. Furthermore, the trajectory of positive affect was independent of that of negative affect and related with distinct goal adjustment processes (i.e., goal disengagement vs. goal re-engagement). This finding indicates the need to tailor psychological care to the nature of the adjustment problem.
INTRODUCTION
Colorectal cancer (CRC) is not only the third most common cancer type worldwide, but is also among the cancer types with the best survival rates (American Cancer Society, 2016). As a result, a large number of individuals is living with the consequences of CRC, which may differ per phase in the disease continuum. While the active treatment phase confronts individuals with invasive treatments and their side effects, the recovery phase may present challenges related to resuming daily activities in light of physical limitations. Accordingly, there has been considerable attention for how people adjust to these various challenges over time (Stanton, Revenson, & Tennen, 2007).
Informing this question, studies have examined how individuals differ in their trajectory of various adjustment indices after colorectal cancer diagnosis (Dunn, Ng, Holland et al., 2013; Dunn, Ng, Breitbart et al., 2013; Hou, Law, Yin, & Fu, 2010). For example, a study investigating trajectories of distress from five months to five years post diagnosis found that while some patients showed low (19.4%) or decreasing distress (29.6%), others
demonstrated constant high (12.5%) or increasing distress over time (38.5%) (Dunn et al., 2013). Such evidence gives important insight into the proportion of patients that may benefit from psychosocial care, but also in the appropriate timing of this care. For instance, it suggests that a subgroup of patients may benefit from care later along the disease continuum, rather than shortly after diagnosis. Especially in a highly prevalent condition such as CRC, such information is essential for the allocation of scarce resources.
Up to date, however, studies in CRC as well as other cancer types have typically focused on psychological distress (e.g., anxiety, depression) as indicator of adjustment, while little attention has been paid to positive and negative affect. Yet, insight into the trajectory of positive and negative affect after cancer diagnosis may help to better
understand the nature of adjustment problems. For instance, a study found that a sample of patients with advanced cancer only differed from the general population in their levels of positive affect (e.g., joy, contention, pleasure) but not in levels of negative affect (Voogt et al., 2005). This suggests a need to focus more on the absence of positive feelings in the screening and treatment of adjustment problems However, insight into individual
differences in positive and negative affect across the cancer continuum is currently lacking. Hence, the primary aim of the current study was to examine trajectories of affect in a sample of individuals diagnosed with colorectal cancer.
Although positive and negative affect are generally seen as independent dimensions (Watson, Clark, & Tellegen, 1988), literature has also pointed towards a protective role of positive affect in the conservation of resources needed to cope adaptively with an adverse life event such as cancer (Fredrickson, 2001). Hence, it could be that patients who follow a trajectory of high positive affect are more likely to adjust well to CRC, and thus experience lower or more rapidly decreasing levels of negative affect over time. Accordingly, a study demonstrated that stable or increasing positive affect was associated with lower anxiety and
ABSTRACT
Objective:Insight into trajectories of positive and negative affect across the cancer continuum may improve understanding of the nature of adjustment problems. The primary aim of this study was to identify subgroups of patients with distinct trajectories of positive and negative affect following diagnosis of colorectal cancer. Secondary to this aim, the co-occurrence between trajectories and their association with goal-related processes was explored.
Methods: Colorectal cancer patients (n=186) completed questionnaires within one month,
7 months, and 18 months after diagnosis. Multilevel models were used to study the trajectory of positive and negative affect, as measured with the PANAS.
Results: Four classes with distinct positive affect trajectories were identified: low (18.8%),
increasing (6.7%), moderate (68.2%) and high (6.3%); two trajectories of negative affect
emerged: low (36.3%) and moderate (63.7%). There was no significant association between positive and negative trajectory class probabilities. The average trajectory of positive affect co-varied with levels of goal disturbance and goal re-engagement over time, while the average negative affect trajectory co-varied with goal disturbance and goal disengagement.
Conclusions: Compared to the general population, our sample of cancer patients suffered
from a lack of positive emotions, but not a high presence of negative emotions. About one-fifth of patients reported low positive affect up to 18 months after diagnosis and may benefit from supportive care. Furthermore, the trajectory of positive affect was independent of that of negative affect and related with distinct goal adjustment processes (i.e., goal disengagement vs. goal re-engagement). This finding indicates the need to tailor psychological care to the nature of the adjustment problem.
Positive and negative affect after colorectal cancer diagnosis _________________________________________________________________________
INTRODUCTION
Colorectal cancer (CRC) is not only the third most common cancer type worldwide, but is also among the cancer types with the best survival rates (American Cancer Society, 2016). As a result, a large number of individuals is living with the consequences of CRC, which may differ per phase in the disease continuum. While the active treatment phase confronts individuals with invasive treatments and their side effects, the recovery phase may present challenges related to resuming daily activities in light of physical limitations. Accordingly, there has been considerable attention for how people adjust to these various challenges over time (Stanton, Revenson, & Tennen, 2007).
Informing this question, studies have examined how individuals differ in their trajectory of various adjustment indices after colorectal cancer diagnosis (Dunn, Ng, Holland et al., 2013; Dunn, Ng, Breitbart et al., 2013; Hou, Law, Yin, & Fu, 2010). For example, a study investigating trajectories of distress from five months to five years post diagnosis found that while some patients showed low (19.4%) or decreasing distress (29.6%), others
demonstrated constant high (12.5%) or increasing distress over time (38.5%) (Dunn et al., 2013). Such evidence gives important insight into the proportion of patients that may benefit from psychosocial care, but also in the appropriate timing of this care. For instance, it suggests that a subgroup of patients may benefit from care later along the disease continuum, rather than shortly after diagnosis. Especially in a highly prevalent condition such as CRC, such information is essential for the allocation of scarce resources.
Up to date, however, studies in CRC as well as other cancer types have typically focused on psychological distress (e.g., anxiety, depression) as indicator of adjustment, while little attention has been paid to positive and negative affect. Yet, insight into the trajectory of positive and negative affect after cancer diagnosis may help to better
understand the nature of adjustment problems. For instance, a study found that a sample of patients with advanced cancer only differed from the general population in their levels of positive affect (e.g., joy, contention, pleasure) but not in levels of negative affect (Voogt et al., 2005). This suggests a need to focus more on the absence of positive feelings in the screening and treatment of adjustment problems However, insight into individual
differences in positive and negative affect across the cancer continuum is currently lacking. Hence, the primary aim of the current study was to examine trajectories of affect in a sample of individuals diagnosed with colorectal cancer.
Although positive and negative affect are generally seen as independent dimensions (Watson, Clark, & Tellegen, 1988), literature has also pointed towards a protective role of positive affect in the conservation of resources needed to cope adaptively with an adverse life event such as cancer (Fredrickson, 2001). Hence, it could be that patients who follow a trajectory of high positive affect are more likely to adjust well to CRC, and thus experience lower or more rapidly decreasing levels of negative affect over time. Accordingly, a study demonstrated that stable or increasing positive affect was associated with lower anxiety and
depression at three months after colorectal cancer diagnosis (Hou, Law, & Fu, 2010). This finding suggests that intervening upon positive affect may also benefit other outcomes such as distress, however, first more insight is needed on how positive and negative affect actually relate. A secondary aim of this study was therefore to explore the co-occurrence of trajectories of positive and negative affect.
Finally, two processes that may influence the trajectory of positive and negative affect are the disturbance and adjustment of personal goals. Across the cancer continuum, the pursuit of personal goals may be disturbed by disease and treatment-related factors such as physical limitations or time spent in the hospital. Self-regulation theories suggest that such disturbances may not only give rise to negative affect, but also limit positive affect (Carver & Scheier, 1982; Emmons, 1996). In contrast, it has been suggested that goal adjustment plays a protective role in the adaptation to adverse situations such as the confrontation with illness. That is, the disengagement from unattainable goals is expected to reduce the impact of the situation on negative indicators of well-being (e.g., negative affect), while the reengagement in other meaningful goals is expected to promote positive well-being (e.g., positive affect) (Wrosch, Scheier, & Miller, 2013).
Importantly, if these goal disturbance and adjustment processes indeed affect the trajectory of positive and negative affect, they may act as intervention targets for supportive care services offered to individuals at risk of poor adjustment. Two studies in breast cancer have found no effect of goal disturbance on changes in negative affect (Low & Stanton, 2015; Stefanic, Iverson, Caputi, & Lane, 2016), of which one study did find that individuals who reported greater disturbance of concrete goals (i.e., activities) after diagnosis showed a greater decrease in positive affect at 3 month follow-up (Low & Stanton, 2015). Again in the context of breast cancer, two studies found that goal disengagement predicted changes in negative affect (Lam et al., 2015; Thompson, Stanton, & Bower, 2013), while two other studies did not (Mens & Scheier, 2015; Wrosch & Sabiston, 2013). Goal re-engagement has been associated with changes in positive affect in two studies (Mens & Scheier, 2015; Wrosch & Sabiston, 2013), but not in another (Lam et al., 2015).
Thus, although previous studies have provided some support for the impact of goal disturbance and adjustment on the trajectory of positive and negative affect, results are inconsistent and evidence in the specific context of colorectal cancer is lacking. Moreover, previous studies have examined goal disturbance and adjustment as baseline variables, while a recent study suggest that a significant proportion of patients demonstrates increases in goal adjustment over time (Zhu et al., 2015). On the other hand, previous analysis of the present study’s data showed that levels of goal disturbance decrease over time (Janse, Ranchor, Smink, Sprangers, & Fleer, 2015). The final aim of this study was therefore to examine whether these changes in goal disturbance and adjustment co-vary with trajectories of positive and negative affect.
METHODS
Study design and population
Data for this study were collected in the context of a longitudinal observational study in patients with colorectal cancer. More information about the study procedure and participants is published elsewhere (Janse, Fleer, Smink, Sprangers, & Ranchor, 2015; Janse, Sprangers, Ranchor, & Fleer, 2015). Assessments were scheduled within one month (T1), 7 months (T2) and 18 months post diagnosis (T3). These time points were chosen to reflect meaningful periods in the cancer trajectory, namely active treatment (T1-T2) and recovery/return to functioning (T2-T3). Approval for the study was obtained from the Medical Ethical Committee of the University Medical Centre Groningen.
Patients were recruited between September 2011 and March 2013 from four hospitals in the Netherlands. Eligible patients were 18 years or older, newly diagnosed, and did not have cognitive impairment, language problems or an addiction. Of the 622 patients who were eligible for participation, 228 provided informed consent (response rate 45.9%). Of these, 186 patients completed all three assessments and were included in the current study (drop out = 15.1%). Due to ethical considerations, patients who did not provide informed consent did not disclose any personal information and could thus not be compared to those who were included in the study. Patients with incomplete assessments (N=33) did not differ from those who completed all assessments with respect to age, gender, education level, relationship status, disease stage, diagnosis, type of treatment, presence of a stoma or positive affect at T1. However, they did have higher negative affect at T1 (p=.043).
The average age of the study’s sample (N=186) was 64.2 years (SD=10.8, range= 38-93). The majority was male (60.8%), completed medium level education (45.2%) or higher education (37.1%) and was in a relationship (81.7%). Furthermore, the majority of patients was classified into stage I or II (55.9%), diagnosed with a tumor in the colon or sigmoid (58.6%), and received chemotherapy or radiotherapy in addition to surgery (54.3%). Almost one-third (30.6%) of patients had a temporary or permanent stoma at 7-month follow-up.
Procedure
A nurse or physician informed eligible patients about the study and handed them an information package with an information letter, informed consent form and prepaid
envelope. Patients were asked to read the information letter and return the informed consent form if they were willing to participate. At all three timepoints, self-report questionnaires were administered by a trained research assistant at the patient’s home or in the hospital.
Positive and negative affect after colorectal cancer diagnosis
_________________________________________________________________________ depression at three months after colorectal cancer diagnosis (Hou, Law, & Fu, 2010). This finding suggests that intervening upon positive affect may also benefit other outcomes such as distress, however, first more insight is needed on how positive and negative affect actually relate. A secondary aim of this study was therefore to explore the co-occurrence of trajectories of positive and negative affect.
Finally, two processes that may influence the trajectory of positive and negative affect are the disturbance and adjustment of personal goals. Across the cancer continuum, the pursuit of personal goals may be disturbed by disease and treatment-related factors such as physical limitations or time spent in the hospital. Self-regulation theories suggest that such disturbances may not only give rise to negative affect, but also limit positive affect (Carver & Scheier, 1982; Emmons, 1996). In contrast, it has been suggested that goal adjustment plays a protective role in the adaptation to adverse situations such as the confrontation with illness. That is, the disengagement from unattainable goals is expected to reduce the impact of the situation on negative indicators of well-being (e.g., negative affect), while the reengagement in other meaningful goals is expected to promote positive well-being (e.g., positive affect) (Wrosch, Scheier, & Miller, 2013).
Importantly, if these goal disturbance and adjustment processes indeed affect the trajectory of positive and negative affect, they may act as intervention targets for supportive care services offered to individuals at risk of poor adjustment. Two studies in breast cancer have found no effect of goal disturbance on changes in negative affect (Low & Stanton, 2015; Stefanic, Iverson, Caputi, & Lane, 2016), of which one study did find that individuals who reported greater disturbance of concrete goals (i.e., activities) after diagnosis showed a greater decrease in positive affect at 3 month follow-up (Low & Stanton, 2015). Again in the context of breast cancer, two studies found that goal disengagement predicted changes in negative affect (Lam et al., 2015; Thompson, Stanton, & Bower, 2013), while two other studies did not (Mens & Scheier, 2015; Wrosch & Sabiston, 2013). Goal re-engagement has been associated with changes in positive affect in two studies (Mens & Scheier, 2015; Wrosch & Sabiston, 2013), but not in another (Lam et al., 2015).
Thus, although previous studies have provided some support for the impact of goal disturbance and adjustment on the trajectory of positive and negative affect, results are inconsistent and evidence in the specific context of colorectal cancer is lacking. Moreover, previous studies have examined goal disturbance and adjustment as baseline variables, while a recent study suggest that a significant proportion of patients demonstrates increases in goal adjustment over time (Zhu et al., 2015). On the other hand, previous analysis of the present study’s data showed that levels of goal disturbance decrease over time (Janse, Ranchor, Smink, Sprangers, & Fleer, 2015). The final aim of this study was therefore to examine whether these changes in goal disturbance and adjustment co-vary with trajectories of positive and negative affect.
Positive and negative affect after colorectal cancer diagnosis _________________________________________________________________________
METHODS
Study design and population
Data for this study were collected in the context of a longitudinal observational study in patients with colorectal cancer. More information about the study procedure and participants is published elsewhere (Janse, Fleer, Smink, Sprangers, & Ranchor, 2015; Janse, Sprangers, Ranchor, & Fleer, 2015). Assessments were scheduled within one month (T1), 7 months (T2) and 18 months post diagnosis (T3). These time points were chosen to reflect meaningful periods in the cancer trajectory, namely active treatment (T1-T2) and recovery/return to functioning (T2-T3). Approval for the study was obtained from the Medical Ethical Committee of the University Medical Centre Groningen.
Patients were recruited between September 2011 and March 2013 from four hospitals in the Netherlands. Eligible patients were 18 years or older, newly diagnosed, and did not have cognitive impairment, language problems or an addiction. Of the 622 patients who were eligible for participation, 228 provided informed consent (response rate 45.9%). Of these, 186 patients completed all three assessments and were included in the current study (drop out = 15.1%). Due to ethical considerations, patients who did not provide informed consent did not disclose any personal information and could thus not be compared to those who were included in the study. Patients with incomplete assessments (N=33) did not differ from those who completed all assessments with respect to age, gender, education level, relationship status, disease stage, diagnosis, type of treatment, presence of a stoma or positive affect at T1. However, they did have higher negative affect at T1 (p=.043).
The average age of the study’s sample (N=186) was 64.2 years (SD=10.8, range= 38-93). The majority was male (60.8%), completed medium level education (45.2%) or higher education (37.1%) and was in a relationship (81.7%). Furthermore, the majority of patients was classified into stage I or II (55.9%), diagnosed with a tumor in the colon or sigmoid (58.6%), and received chemotherapy or radiotherapy in addition to surgery (54.3%). Almost one-third (30.6%) of patients had a temporary or permanent stoma at 7-month follow-up.
Procedure
A nurse or physician informed eligible patients about the study and handed them an information package with an information letter, informed consent form and prepaid
envelope. Patients were asked to read the information letter and return the informed consent form if they were willing to participate. At all three timepoints, self-report questionnaires were administered by a trained research assistant at the patient’s home or in the hospital.
Measures
Demographic characteristics - Age, gender, education level, and relationship status
were obtained by self-report at the first assessment.
Clinical characteristics - Information on clinical characteristics was obtained from The
Netherlands Cancer Registry (NCR). Dichotomous variables were created for disease stage (I-II versus III-IV) (Edge & Compton, 2010), diagnosis (tumour in colon/sigmoid versus rectum/anus), treatment (surgery only vs. surgery plus adjuvant treatment) and the presence of a stoma (yes vs. no).
Positive and negative affect - Positive and negative affect were measured with the
Positive and Negative Affect Schedule (PANAS) (Watson et al., 1988) at all three timepoints. The PANAS contains two subscales that measured the presence of positive affect (e.g. interested, excited) and negative affect (e.g.,irritable, afraid) in the past week on a scale of 1 ‘very slightly or not at all’ to 5 ‘extremely’. A sum score was obtained for each 10-item subscale, thus, total scores ranged from 10 to 50. Cronbach’s alpha for the positive affect subscale was .84 at T1, .91 at T2, and .88 at T3. For the negative affect subscale, Cronbach’s alpha was .87 at T1, .88 at T2 and .88 at T3.
Goal disturbance - The extent to which goals were disturbed by cancer was assessed
with a nomothetic-idiographic method (Emmons, 1999). At every assessment, patients were asked to list three to ten personal goals, which were explained as the plans/projects they are currently working on and are aimed at the things they want to attain, maintain or avoid. For each goal, patients rated the extent to which they perceived their goal to be disturbed by cancer on a scale of 1 ‘not at all’ to 10 ‘very’. The total score for goal disturbance represents the mean rating across an individual’s goals.
Goal adjustment - Goal adjustment was measured with The Goal Disengagement and
Reengagement Scale (GDRS) (Wrosch, Scheier, Miller, Schulz, & Carver, 2003). The GDRS consists of 10 items of which four measure goal disengagement capacities (e.g., “If I have to stop pursuing an important goal in my life, it’s easy for me to reduce my effort toward the goals”) and six measure goal re-engagement capacities (e.g., “If I have to stop pursuing an important goal in my life, I seek other meaningful goals). Patients were asked to respond to these items on a scale ranging from 1 ‘almost never true’ to 5 ‘almost always true’. Sum scores were calculated for both subscales. Cronbach’s alpha for the
disengagement scale was .63 at T1, .76 at T2, and .74 at T3. For the re-engagement scale, this was .91 at T1, .91 at T2, and .92 at T3.
Statistical analyses
Research questions were examined using multilevel modelling, which allowed us to examine the trajectory of positive and negative affect across the three timepoints and control for dependency in the data due to repeated measurements. In all models, time was
considered as categorical (dummy coding) – thus, no a priori shape of the trajectory was assumed.
First, Latent Class Growth Analysis (LCGA), a group-based semi-parametric approach based on finite mixture modeling (Nagin, 2005), was used to examine how patients differed in their trajectory of positive and negative affect over time. LCGA summarizes longitudinal data by modeling individual-level heterogeneity through a small number of homogeneous classes of trajectories (Jung & Wickrama, 2008; Nagin & Odgers, 2010). LCGA analyses were conducted with Mplus version 7.1 (Muthen & Muthen, 2012) and LatentGold version 4.5 (Vermunt & Magidson, 2010). Separate LCGA models were examined for positive and negative affect respectively. We tested models that ranged from 1 to 5 classes, with 1000 different starting values to avoid local-optimum estimation.
Several criteria were used to determine the number of classes that best described patterns of change in the data. The Bayesian Information Criterion (BIC), adjusted BIC, and Akaike’s Information Criterion (AIC) were used as a measure of model fit – lower values indicate better model fit (Nylund, Asparouhov, & Muthén, 2007). Entropy values of .6 or higher indicate adequate class separation (Nylund et al., 2007). When fit indices did not unanimously identify one best model, we based our final decision on conceptual meaningfulness of the models. A class was only considered meaningful when it showed a clearly different pattern of change compared to the other classes and contained a substantial (>5%) number of patients.
Second, we tested a dual trajectory model (Nagin, 2005), in which the bivariate class membership probabilities for both positive and negative affect trajectories were estimated. Model parameters from the previously estimated trajectories where used as starting values. The joint (bivariate) membership probabilities captures the overlap across both set of trajectories. To examine whether positive and negative affect trajectory classes were indeed associated, we compared a model in which class probabilities were related to a model in which they were independent – by means of the fit indices (AIC and BIC), as well as the significance of the parameters that relates both sets of class membership probabilities.
Third, conventional multilevel models were estimated to examine the covariance between goal disturbance, goal adjustment, and affect, using affect variables as outcome and goal variables as time-varying covariates. As we were interested in the average effect of these variables, we examined this for the sample as a whole rather than per trajectory class. A random intercept was included to control for non-independence of residuals. Predictors were grand-mean centered to facilitate interpretation of their effects. Demographic and clinical variables were included as potential confounders when they demonstrated a significant univariate association (p<.05) with both the outcome variable and at least one of the predictors.
Positive and negative affect after colorectal cancer diagnosis
_________________________________________________________________________
Measures
Demographic characteristics - Age, gender, education level, and relationship status
were obtained by self-report at the first assessment.
Clinical characteristics - Information on clinical characteristics was obtained from The
Netherlands Cancer Registry (NCR). Dichotomous variables were created for disease stage (I-II versus III-IV) (Edge & Compton, 2010), diagnosis (tumour in colon/sigmoid versus rectum/anus), treatment (surgery only vs. surgery plus adjuvant treatment) and the presence of a stoma (yes vs. no).
Positive and negative affect - Positive and negative affect were measured with the
Positive and Negative Affect Schedule (PANAS) (Watson et al., 1988) at all three timepoints. The PANAS contains two subscales that measured the presence of positive affect (e.g. interested, excited) and negative affect (e.g.,irritable, afraid) in the past week on a scale of 1 ‘very slightly or not at all’ to 5 ‘extremely’. A sum score was obtained for each 10-item subscale, thus, total scores ranged from 10 to 50. Cronbach’s alpha for the positive affect subscale was .84 at T1, .91 at T2, and .88 at T3. For the negative affect subscale, Cronbach’s alpha was .87 at T1, .88 at T2 and .88 at T3.
Goal disturbance - The extent to which goals were disturbed by cancer was assessed
with a nomothetic-idiographic method (Emmons, 1999). At every assessment, patients were asked to list three to ten personal goals, which were explained as the plans/projects they are currently working on and are aimed at the things they want to attain, maintain or avoid. For each goal, patients rated the extent to which they perceived their goal to be disturbed by cancer on a scale of 1 ‘not at all’ to 10 ‘very’. The total score for goal disturbance represents the mean rating across an individual’s goals.
Goal adjustment - Goal adjustment was measured with The Goal Disengagement and
Reengagement Scale (GDRS) (Wrosch, Scheier, Miller, Schulz, & Carver, 2003). The GDRS consists of 10 items of which four measure goal disengagement capacities (e.g., “If I have to stop pursuing an important goal in my life, it’s easy for me to reduce my effort toward the goals”) and six measure goal re-engagement capacities (e.g., “If I have to stop pursuing an important goal in my life, I seek other meaningful goals). Patients were asked to respond to these items on a scale ranging from 1 ‘almost never true’ to 5 ‘almost always true’. Sum scores were calculated for both subscales. Cronbach’s alpha for the
disengagement scale was .63 at T1, .76 at T2, and .74 at T3. For the re-engagement scale, this was .91 at T1, .91 at T2, and .92 at T3.
Statistical analyses
Research questions were examined using multilevel modelling, which allowed us to examine the trajectory of positive and negative affect across the three timepoints and control for dependency in the data due to repeated measurements. In all models, time was
Positive and negative affect after colorectal cancer diagnosis _________________________________________________________________________ considered as categorical (dummy coding) – thus, no a priori shape of the trajectory was assumed.
First, Latent Class Growth Analysis (LCGA), a group-based semi-parametric approach based on finite mixture modeling (Nagin, 2005), was used to examine how patients differed in their trajectory of positive and negative affect over time. LCGA summarizes longitudinal data by modeling individual-level heterogeneity through a small number of homogeneous classes of trajectories (Jung & Wickrama, 2008; Nagin & Odgers, 2010). LCGA analyses were conducted with Mplus version 7.1 (Muthen & Muthen, 2012) and LatentGold version 4.5 (Vermunt & Magidson, 2010). Separate LCGA models were examined for positive and negative affect respectively. We tested models that ranged from 1 to 5 classes, with 1000 different starting values to avoid local-optimum estimation.
Several criteria were used to determine the number of classes that best described patterns of change in the data. The Bayesian Information Criterion (BIC), adjusted BIC, and Akaike’s Information Criterion (AIC) were used as a measure of model fit – lower values indicate better model fit (Nylund, Asparouhov, & Muthén, 2007). Entropy values of .6 or higher indicate adequate class separation (Nylund et al., 2007). When fit indices did not unanimously identify one best model, we based our final decision on conceptual meaningfulness of the models. A class was only considered meaningful when it showed a clearly different pattern of change compared to the other classes and contained a substantial (>5%) number of patients.
Second, we tested a dual trajectory model (Nagin, 2005), in which the bivariate class membership probabilities for both positive and negative affect trajectories were estimated. Model parameters from the previously estimated trajectories where used as starting values. The joint (bivariate) membership probabilities captures the overlap across both set of trajectories. To examine whether positive and negative affect trajectory classes were indeed associated, we compared a model in which class probabilities were related to a model in which they were independent – by means of the fit indices (AIC and BIC), as well as the significance of the parameters that relates both sets of class membership probabilities.
Third, conventional multilevel models were estimated to examine the covariance between goal disturbance, goal adjustment, and affect, using affect variables as outcome and goal variables as time-varying covariates. As we were interested in the average effect of these variables, we examined this for the sample as a whole rather than per trajectory class. A random intercept was included to control for non-independence of residuals. Predictors were grand-mean centered to facilitate interpretation of their effects. Demographic and clinical variables were included as potential confounders when they demonstrated a significant univariate association (p<.05) with both the outcome variable and at least one of the predictors.
RESULTS Assumptions
Multilevel models examining the trajectory of positive and negative affect across the three time-points included two levels, i.e., assessments nested within individuals, resulting in a total of 558 observations. Negative skewness was observed for negative affect, but since transformation did not improve distribution this variable remained untransformed. There were no extreme univariate or multivariate outliers (p<.001). Two cases did not complete affect measures at T3, however, since data were missing at random these cases were included in the analyses.
Table 1 Model selection Latent Class Growth Analysis (LCGA)
Individual differences in the trajectory of positive and negative affect
Latent Class Growth Models were used to identify latent classes of individuals with a distinct trajectory of positive and negative affect.
Positive affect – As shown in table 1, the BIC criteria suggested that a three-class model
was preferable, whereas the AIC and adjusted BIC criteria favored a five-class model. On the other hand, the four-class model showed the best class separation. Hence, the final decision was based on non-statistical criteria. Compared to the three-class model, the four-class model showed a small extra four-class with a considerable increase in positive affect from T1 to T2. This class still contained a satisfactory proportion of patients (6.3%) and showed a pattern that seemed both clinically meaningful and substantially different from that of the other classes. On the other hand, the five-class model showed only a very small extra class
(3.2%) which did not demonstrate a pattern that was substantially different from the other classes. Thus, a four-class model was considered best.
Trajectories of positive affect for the four identified classes are shown in figure 1. In addition, table 2 presents mean levels of positive affect as well as effect sizes for the change in affect between time points within each trajectory class. The low positive affect class (18.8%) showed a moderate decrease between T1 and T2, but bounced back to
post-diagnosis levels between T2 and T3. The increasing positive affect class (6.7%) started with low positive affect but demonstrated a large increase between T1 and T2 and further improved between T2 and T3. The largest class moderate positive affect (68.2%) showed a small increase in positive affect from T1 and T2 and stable levels thereafter (T2-T3). Finally, the high positive affect class (6.3%) demonstrated a large increase from T1 to T2 and a small decrease between T2 and T3.
Negative affect – AIC and BIC criteria favored the three-class model (see table 1).
However, compared to a two-class model, this model identified a class of patients with constant values of 10 over time. As this was the lowest possible value, it seemed that a floor effect caused the software to separate these patients from others with low negative affect score. However, as the other class with low scores did not have substantially higher scores than these patients, it did not seem conceptually meaningful to separate these two classes. Hence, we decided to merge these two classes, resulting in a final two-class solution.
Figure 2 presents the trajectories of the two identified classes. As shown in table 2, the
low negative affect class (36.3%) showed a small decrease in negative affect across the total
follow-up period. The moderate negative affect class (63.7%) showed a small decrease in the first follow-up period (T1-T2) and stable levels in the period thereafter (T2-T3).
Table 2 Average positive and negative affect at T1, T2 and T3, and magnitude of the
Positive and negative affect after colorectal cancer diagnosis
_________________________________________________________________________
RESULTS Assumptions
Multilevel models examining the trajectory of positive and negative affect across the three time-points included two levels, i.e., assessments nested within individuals, resulting in a total of 558 observations. Negative skewness was observed for negative affect, but since transformation did not improve distribution this variable remained untransformed. There were no extreme univariate or multivariate outliers (p<.001). Two cases did not complete affect measures at T3, however, since data were missing at random these cases were included in the analyses.
Table 1 Model selection Latent Class Growth Analysis (LCGA)
Individual differences in the trajectory of positive and negative affect
Latent Class Growth Models were used to identify latent classes of individuals with a distinct trajectory of positive and negative affect.
Positive affect – As shown in table 1, the BIC criteria suggested that a three-class model
was preferable, whereas the AIC and adjusted BIC criteria favored a five-class model. On the other hand, the four-class model showed the best class separation. Hence, the final decision was based on non-statistical criteria. Compared to the three-class model, the four-class model showed a small extra four-class with a considerable increase in positive affect from T1 to T2. This class still contained a satisfactory proportion of patients (6.3%) and showed a pattern that seemed both clinically meaningful and substantially different from that of the other classes. On the other hand, the five-class model showed only a very small extra class
Positive and negative affect after colorectal cancer diagnosis _________________________________________________________________________ (3.2%) which did not demonstrate a pattern that was substantially different from the other classes. Thus, a four-class model was considered best.
Trajectories of positive affect for the four identified classes are shown in figure 1. In addition, table 2 presents mean levels of positive affect as well as effect sizes for the change in affect between time points within each trajectory class. The low positive affect class (18.8%) showed a moderate decrease between T1 and T2, but bounced back to
post-diagnosis levels between T2 and T3. The increasing positive affect class (6.7%) started with low positive affect but demonstrated a large increase between T1 and T2 and further improved between T2 and T3. The largest class moderate positive affect (68.2%) showed a small increase in positive affect from T1 and T2 and stable levels thereafter (T2-T3). Finally, the high positive affect class (6.3%) demonstrated a large increase from T1 to T2 and a small decrease between T2 and T3.
Negative affect – AIC and BIC criteria favored the three-class model (see table 1).
However, compared to a two-class model, this model identified a class of patients with constant values of 10 over time. As this was the lowest possible value, it seemed that a floor effect caused the software to separate these patients from others with low negative affect score. However, as the other class with low scores did not have substantially higher scores than these patients, it did not seem conceptually meaningful to separate these two classes. Hence, we decided to merge these two classes, resulting in a final two-class solution.
Figure 2 presents the trajectories of the two identified classes. As shown in table 2, the
low negative affect class (36.3%) showed a small decrease in negative affect across the total
follow-up period. The moderate negative affect class (63.7%) showed a small decrease in the first follow-up period (T1-T2) and stable levels in the period thereafter (T2-T3).
Table 2 Average positive and negative affect at T1, T2 and T3, and magnitude of the
Figure 1 Trajectories of positive affect
Figure 2 Trajectories of negative affect
Covariance between trajectories of positive and negative affect
The covariance between trajectory classes of positive and negative affect was examined by comparing two nested models, one in which class probabilities were independent (AIC = 7112.96, BIC = 7145.22) and one in which they were associated (AIC = 7116.73, BIC = 7158.66). Please note that there are no significance tests for comparing AIC and BIC values. However, the differences between the two models were small, suggesting that a model in which trajectory classes were independent was favorable. Furthermore, the interaction parameters that modelled the association were not significant (Wald(0) = 2.05, df=3, p=0.56). Taken together, these findings suggest that positive and negative affect trajectory classes did not co-vary. That is, the probability of following a given trajectory of positive affect was not associated with a higher probability of following a given trajectory of negative affect.
Covariance of goal disturbance and goal adjustment with affect trajectories
Positive affect – A model in which goal disturbance, goal disengagement, and goal
reengagement were included as time-varying predictors of positive affect (PA) explained significant additional variance beyond a model in which only the variability in individuals (intercept) was considered, X2 (df=3) = 3701,687 - 3642,538 = 59.149, p<.001. Age,
education level, and the presence of a stoma were included as potential confounders. However, this did not substantially change the effects of our predictors, suggesting no actual confounding effect.
Table 3 shows that only goal disturbance and goal disengagement were significantly associated with levels of PA over time. However, also the effect of goal disengagement approached significance (p=.053). Goal disturbance was negatively associated with PA; each unit increase in goal disturbance was associated with a .53 decrease in PA on a scale of 10 to 50. On the other hand, goal re-engagement was positively associated with PA. That is, for each unit increase in re-engagement, PA increased with .36.
Negative affect – Including goal disturbance, goal disengagement, and goal
re-engagement as time-varying predictors of negative affect (NA) improved the model compared to a model including only the intercept, X2 (df=3) = 3643,848 - 3588,850 =
54,998, p<.001. Age, presence of a stoma, treatment, and tumor site were included as potential confounders, but did not change the effects of the predictors thus suggesting no confounding effect.
As shown in table 3, goal disturbance and goal disengagement were significantly associated with NA over time, but goal re-engagement was not. On average, goal
disturbance was positively associated with NA; for each unit increase in goal disturbance, NA increased by .48 on a scale of 10 to 50. In contrast, goal disengagement was negatively associated with NA. That is, a unit increase in goal disturbance corresponded with a -.45
Positive and negative affect after colorectal cancer diagnosis
_________________________________________________________________________
Figure 1 Trajectories of positive affect
Figure 2 Trajectories of negative affect
Positive and negative affect after colorectal cancer diagnosis _________________________________________________________________________
Covariance between trajectories of positive and negative affect
The covariance between trajectory classes of positive and negative affect was examined by comparing two nested models, one in which class probabilities were independent (AIC = 7112.96, BIC = 7145.22) and one in which they were associated (AIC = 7116.73, BIC = 7158.66). Please note that there are no significance tests for comparing AIC and BIC values. However, the differences between the two models were small, suggesting that a model in which trajectory classes were independent was favorable. Furthermore, the interaction parameters that modelled the association were not significant (Wald(0) = 2.05, df=3, p=0.56). Taken together, these findings suggest that positive and negative affect trajectory classes did not co-vary. That is, the probability of following a given trajectory of positive affect was not associated with a higher probability of following a given trajectory of negative affect.
Covariance of goal disturbance and goal adjustment with affect trajectories
Positive affect – A model in which goal disturbance, goal disengagement, and goal
reengagement were included as time-varying predictors of positive affect (PA) explained significant additional variance beyond a model in which only the variability in individuals (intercept) was considered, X2 (df=3) = 3701,687 - 3642,538 = 59.149, p<.001. Age,
education level, and the presence of a stoma were included as potential confounders. However, this did not substantially change the effects of our predictors, suggesting no actual confounding effect.
Table 3 shows that only goal disturbance and goal disengagement were significantly associated with levels of PA over time. However, also the effect of goal disengagement approached significance (p=.053). Goal disturbance was negatively associated with PA; each unit increase in goal disturbance was associated with a .53 decrease in PA on a scale of 10 to 50. On the other hand, goal re-engagement was positively associated with PA. That is, for each unit increase in re-engagement, PA increased with .36.
Negative affect – Including goal disturbance, goal disengagement, and goal
re-engagement as time-varying predictors of negative affect (NA) improved the model compared to a model including only the intercept, X2 (df=3) = 3643,848 - 3588,850 =
54,998, p<.001. Age, presence of a stoma, treatment, and tumor site were included as potential confounders, but did not change the effects of the predictors thus suggesting no confounding effect.
As shown in table 3, goal disturbance and goal disengagement were significantly associated with NA over time, but goal re-engagement was not. On average, goal
disturbance was positively associated with NA; for each unit increase in goal disturbance, NA increased by .48 on a scale of 10 to 50. In contrast, goal disengagement was negatively associated with NA. That is, a unit increase in goal disturbance corresponded with a -.45 decrease in NA.
Table 3 Fixed effects of models examining the covariance between goal disturbance, goal
adjustment, and positive and negative affect over time
DISCUSSION
The present study took a comprehensive approach to studying the nature of adjustment to colorectal cancer by examining heterogeneity in trajectories of positive and negative affect from 1 month to 18 months after diagnosis, and processes associated with these trajectories. Four classes with a distinct trajectory of positive affect were identified: low, increasing,
moderate, and high; two classes emerged for negative affect: low and moderate. Positive
and negative affect trajectory classes were found to be independent. Different goal adjustment processes co-varied with the trajectory of positive and negative affect respectively, while both co-varied with levels of goal disturbance.
Except for the ‘high positive affect’ class, levels of positive affect in all other classes were lower than those reported by individuals in a general population sample (Engelen, De Peuter, Victoir, Van Diest, & Van den Bergh, 2006). In contrast, levels of negative affect were on average lower than those found in the general population (Engelen et al., 2006). These findings are in line with those of a previous cross-sectional study in advanced cancer (Voogt et al., 2005) and suggest that cancer patients may predominantly suffer from a lack of positive emotions, rather than a high presence of negative emotions. However, as the PANAS mainly measures high arousal negative emotions (e.g., hostile, upset), it could be that some of the low arousal negative emotions that may arise in the context of cancer (e.g., irritability, restlessness) were not detected (Watson & Tellegen, 1985).
Findings of this study furthermore suggest that lower positive affect does not have consequences for the presence of negative affect. That is, positive and negative affect
trajectory classes were found to be independent, which seems consistent with the broader literature suggesting that positive and negative affect are separate dimensions (Watson et al., 1988). Findings also provide support for theory suggesting that negative emotions predominantly arise from a perceived threat to goals, while positive emotions arise from the engagement in rewarding activities (Carver & White, 1994). Accordingly, we found that the disengagement from unattainable goals co-varied with the trajectory of negative affect and the re-engagement in alternative goals with the trajectory of positive affect.
To our knowledge, this is the first study to investigate goal disturbance and adjustment processes in relation to affect in the specific context of colorectal cancer. Patients with this type of cancer may face unique challenges, such as fatigue or the presence of a stoma. Our findings indicate that the disturbance of personal goals, potentially as a result of such challenges (Janse et al., 2015), can affect both positive and negative aspects of well-being. Although the specific threats to goal pursuit may be diagnosis-specific, the effect of goal disengagement and reengagement was found to be similar to that reported in breast cancer samples (Lam et al., 2015; Low & Stanton, 2015; Mens & Scheier, 2015; Wrosch et al., 2013). The present study further demonstrated that variation in these processes over time can co-vary with levels of affect across the disease continuum, which indicates that intervening upon these processes may attenuate the course of positive and negative affect.
A number of limitations should be acknowledged. First, we had no data of individuals who refused or were unwilling to participate and could therefore not estimate the impact of a selection bias. The Netherlands Cancer Registry reports that 51.3% of Dutch colorectal cancer patients are male and the average age at diagnosis is 67 years, which suggests that our sample was at least fairly representative with respect to age and gender. Second, studies in larger samples may examine whether the effect of goal disturbance and adjustment differs across trajectory classes, as our study lacked power to do so due to the limited number of patients in some of the classes. Third, future research may also include a larger number of assessment and shorter time periods in between assessments to examine the causality of these relationships. Fourth, low reliability of the goal disengagement subscale at baseline in our study and an earlier study (Arends, Bode, Taal, & Laar, 2016) signals a potential problem with the Dutch translation of this scale.
The findings of this study may have implications for clinical practice. First, the loss of positive emotions (e.g., pleasure, interest) is a key feature of depression (Watson, Clark, & Carey, 1988). An earlier study indeed found that colorectal cancer patients who
demonstrated a loss of positive affect after diagnosis reported greater depression at 3 month-follow up, whereas those who gained positive affect reported lower depression (Hou et al., 2010). In addition, a lack of positive affect could hinder the restoration of coping resources (Folkman, 2008; Fredrickson, 2001). Studies have for example found that cancer patients with lower positive affect report poorer social functioning and a lower ability to find meaning (Hirsch, Floyd, & Duberstein, 2012; Voogt et al., 2005).When screening for adjustment problems, it may thus be important to include items measuring a lack of positive
Positive and negative affect after colorectal cancer diagnosis
_________________________________________________________________________
Table 3 Fixed effects of models examining the covariance between goal disturbance, goal
adjustment, and positive and negative affect over time
DISCUSSION
The present study took a comprehensive approach to studying the nature of adjustment to colorectal cancer by examining heterogeneity in trajectories of positive and negative affect from 1 month to 18 months after diagnosis, and processes associated with these trajectories. Four classes with a distinct trajectory of positive affect were identified: low, increasing,
moderate, and high; two classes emerged for negative affect: low and moderate. Positive
and negative affect trajectory classes were found to be independent. Different goal adjustment processes co-varied with the trajectory of positive and negative affect respectively, while both co-varied with levels of goal disturbance.
Except for the ‘high positive affect’ class, levels of positive affect in all other classes were lower than those reported by individuals in a general population sample (Engelen, De Peuter, Victoir, Van Diest, & Van den Bergh, 2006). In contrast, levels of negative affect were on average lower than those found in the general population (Engelen et al., 2006). These findings are in line with those of a previous cross-sectional study in advanced cancer (Voogt et al., 2005) and suggest that cancer patients may predominantly suffer from a lack of positive emotions, rather than a high presence of negative emotions. However, as the PANAS mainly measures high arousal negative emotions (e.g., hostile, upset), it could be that some of the low arousal negative emotions that may arise in the context of cancer (e.g., irritability, restlessness) were not detected (Watson & Tellegen, 1985).
Findings of this study furthermore suggest that lower positive affect does not have consequences for the presence of negative affect. That is, positive and negative affect
Positive and negative affect after colorectal cancer diagnosis _________________________________________________________________________ trajectory classes were found to be independent, which seems consistent with the broader literature suggesting that positive and negative affect are separate dimensions (Watson et al., 1988). Findings also provide support for theory suggesting that negative emotions predominantly arise from a perceived threat to goals, while positive emotions arise from the engagement in rewarding activities (Carver & White, 1994). Accordingly, we found that the disengagement from unattainable goals co-varied with the trajectory of negative affect and the re-engagement in alternative goals with the trajectory of positive affect.
To our knowledge, this is the first study to investigate goal disturbance and adjustment processes in relation to affect in the specific context of colorectal cancer. Patients with this type of cancer may face unique challenges, such as fatigue or the presence of a stoma. Our findings indicate that the disturbance of personal goals, potentially as a result of such challenges (Janse et al., 2015), can affect both positive and negative aspects of well-being. Although the specific threats to goal pursuit may be diagnosis-specific, the effect of goal disengagement and reengagement was found to be similar to that reported in breast cancer samples (Lam et al., 2015; Low & Stanton, 2015; Mens & Scheier, 2015; Wrosch et al., 2013). The present study further demonstrated that variation in these processes over time can co-vary with levels of affect across the disease continuum, which indicates that intervening upon these processes may attenuate the course of positive and negative affect.
A number of limitations should be acknowledged. First, we had no data of individuals who refused or were unwilling to participate and could therefore not estimate the impact of a selection bias. The Netherlands Cancer Registry reports that 51.3% of Dutch colorectal cancer patients are male and the average age at diagnosis is 67 years, which suggests that our sample was at least fairly representative with respect to age and gender. Second, studies in larger samples may examine whether the effect of goal disturbance and adjustment differs across trajectory classes, as our study lacked power to do so due to the limited number of patients in some of the classes. Third, future research may also include a larger number of assessment and shorter time periods in between assessments to examine the causality of these relationships. Fourth, low reliability of the goal disengagement subscale at baseline in our study and an earlier study (Arends, Bode, Taal, & Laar, 2016) signals a potential problem with the Dutch translation of this scale.
The findings of this study may have implications for clinical practice. First, the loss of positive emotions (e.g., pleasure, interest) is a key feature of depression (Watson, Clark, & Carey, 1988). An earlier study indeed found that colorectal cancer patients who
demonstrated a loss of positive affect after diagnosis reported greater depression at 3 month-follow up, whereas those who gained positive affect reported lower depression (Hou et al., 2010). In addition, a lack of positive affect could hinder the restoration of coping resources (Folkman, 2008; Fredrickson, 2001). Studies have for example found that cancer patients with lower positive affect report poorer social functioning and a lower ability to find meaning (Hirsch, Floyd, & Duberstein, 2012; Voogt et al., 2005).When screening for adjustment problems, it may thus be important to include items measuring a lack of positive
emotions. As this study found that positive affect stabilizes after active treatment, the recovery period seems a relevant period to screen for this outcome.
Our results further indicated that about one-fifth (18.8%) of patients has low positive affect up to 18 months after diagnosis and may potentially benefit from psychosocial support. Findings also suggest that interventions offered to this group of patients may need to target different processes than interventions for patients with high negative affect. Although all patients may benefit from setting attainable goals, patients with low positive affect may in particular need assistance with identifying rewarding activities that can be pursued despite disease-related limitations. On the other hand, patients with high negative affect may benefit from strategies to disengage from unattainable goals, for instance by scaling them down. Although both disengagement and re-engagement strategies are typically included in acceptance- or mindfulness-based cognitive therapies, tailoring these strategies may help to achieve optimal results.
References
American Cancer Society. (2016). Cancer facts and figures. Atlanta: American Cancer Society.
Arends, R. Y., Bode, C., Taal, E., & Laar, M. A. (2016). The longitudinal relation between patterns of goal management and psychological health in people with arthritis: The need for adaptive flexibility. British Journal of Health Psychology, 21, 469 - 498.
Carver, C. S., & Scheier, M. F. (1982). Control theory: A useful conceptual framework for personality–social, clinical, and health psychology. Psychological Bulletin, 92(1), 111 - 135.
Carver, C. S., & White, T. L. (1994). Behavioral-inhibition, behavioral activation, and affective responses to impending reward and punishment - the BIS BAS scales. Journal of Personality and Social Psychology, 67(2), 319-333.
Dunn, J., Ng, S. K., Breitbart, W., Aitken, J., Youl, P., Baade, P. D., & Chambers, S. K. (2013). Health-related quality of life and life satisfaction in colorectal cancer survivors: Trajectories of adjustment. Health and Quality of Life Outcomes, 11, 46.
Dunn, J., Ng, S. K., Holland, J., Aitken, J., Youl, P., Baade, P. D., & Chambers, S. K. (2013). Trajectories of psychological distress after colorectal cancer. Psycho-Oncology, 22(8), 1759-1765.
Edge, S. B., & Compton, C. C. (2010). The American joint committee on cancer: The 7th edition of the AJCC cancer staging manual and the future of TNM. Annals of Surgical Oncology, 17(6), 1471-1474.
Emmons, R. A. (1996). Striving and feeling: Personal goals and subjective well-being. In P. M. Gollwitzer, & J. A. Bargh (Eds.), The psychology of action: Linking cognition and motivation to behavior (pp. 313-337). New York, NY: Guilford Press.
Emmons, R. A. (1999). The psychology of ultimate concerns: Motivation and spirituality in personality. New York, NY: Guilford Press.
Engelen, U., De Peuter, S., Victoir, A., Van Diest, I., & Van den Bergh, O. (2006). Verdere validering van de positive and negative affect schedule (PANAS) en vergelijking van twee nederlandstalige versies. Gedrag En Gezondheid, 34(2), 61-70.
Folkman, S. (2008). The case for positive emotions in the stress process. Anxiety Stress and Coping, 21(1), 3-14. Fredrickson, B. (2001). The role of positive emotions in positive psychology - the broaden-and-build theory of
positive emotions. American Psychologist, 56(3), 218-226.
Hirsch, J. K., Floyd, A. R., & Duberstein, P. R. (2012). Perceived health in lung cancer patients: The role of positive and negative affect. Quality of Life Research, 21(2), 187-194.
Hou, W. K., Law, C. C., & Fu, Y. T. (2010). Does change in positive affect mediate and/or moderate the impact of symptom distress on psychological adjustment after cancer diagnosis? A prospective analysis. Psychology & Health, 25(4), 417-431.
Hou, W. K., Law, C. C., Yin, J., & Fu, Y. T. (2010). Resource loss, resource gain, and psychological resilience and dysfunction following cancer diagnosis: A growth mixture modeling approach. Health Psychology, 29(5), 484-495.
Janse, M., Fleer, J., Smink, A., Sprangers, M. A., & Ranchor, A. V. (2015). Which goal adjustment strategies do cancer patients use? A longitudinal study. Psycho‐Oncology, 25(3), 332-338.
Janse, M., Sprangers, M. A., Ranchor, A. V., & Fleer, J. (2015). Long-term effects of goal disturbance and adjustment on well-being in cancer patients. Quality of Life Research, 25(4), 1017-1027.
Janse, M., Ranchor, A. V., Smink, A., Sprangers, M. A. G., & Fleer, J. (2015). Changes in cancer patients' personal goals in the first 6 months after diagnosis: The role of illness variables. Supportive Care in Cancer: Official Journal of the Multinational Association of Supportive Care in Cancer, 23(7), 1893-1900. Jung, T., & Wickrama, K. (2008). An introduction to latent class growth analysis and growth mixture modeling.
Positive and negative affect after colorectal cancer diagnosis
_________________________________________________________________________ emotions. As this study found that positive affect stabilizes after active treatment, the recovery period seems a relevant period to screen for this outcome.
Our results further indicated that about one-fifth (18.8%) of patients has low positive affect up to 18 months after diagnosis and may potentially benefit from psychosocial support. Findings also suggest that interventions offered to this group of patients may need to target different processes than interventions for patients with high negative affect. Although all patients may benefit from setting attainable goals, patients with low positive affect may in particular need assistance with identifying rewarding activities that can be pursued despite disease-related limitations. On the other hand, patients with high negative affect may benefit from strategies to disengage from unattainable goals, for instance by scaling them down. Although both disengagement and re-engagement strategies are typically included in acceptance- or mindfulness-based cognitive therapies, tailoring these strategies may help to achieve optimal results.
Positive and negative affect after colorectal cancer diagnosis _________________________________________________________________________
References
American Cancer Society. (2016). Cancer facts and figures. Atlanta: American Cancer Society.
Arends, R. Y., Bode, C., Taal, E., & Laar, M. A. (2016). The longitudinal relation between patterns of goal management and psychological health in people with arthritis: The need for adaptive flexibility. British Journal of Health Psychology, 21, 469 - 498.
Carver, C. S., & Scheier, M. F. (1982). Control theory: A useful conceptual framework for personality–social, clinical, and health psychology. Psychological Bulletin, 92(1), 111 - 135.
Carver, C. S., & White, T. L. (1994). Behavioral-inhibition, behavioral activation, and affective responses to impending reward and punishment - the BIS BAS scales. Journal of Personality and Social Psychology, 67(2), 319-333.
Dunn, J., Ng, S. K., Breitbart, W., Aitken, J., Youl, P., Baade, P. D., & Chambers, S. K. (2013). Health-related quality of life and life satisfaction in colorectal cancer survivors: Trajectories of adjustment. Health and Quality of Life Outcomes, 11, 46.
Dunn, J., Ng, S. K., Holland, J., Aitken, J., Youl, P., Baade, P. D., & Chambers, S. K. (2013). Trajectories of psychological distress after colorectal cancer. Psycho-Oncology, 22(8), 1759-1765.
Edge, S. B., & Compton, C. C. (2010). The American joint committee on cancer: The 7th edition of the AJCC cancer staging manual and the future of TNM. Annals of Surgical Oncology, 17(6), 1471-1474.
Emmons, R. A. (1996). Striving and feeling: Personal goals and subjective well-being. In P. M. Gollwitzer, & J. A. Bargh (Eds.), The psychology of action: Linking cognition and motivation to behavior (pp. 313-337). New York, NY: Guilford Press.
Emmons, R. A. (1999). The psychology of ultimate concerns: Motivation and spirituality in personality. New York, NY: Guilford Press.
Engelen, U., De Peuter, S., Victoir, A., Van Diest, I., & Van den Bergh, O. (2006). Verdere validering van de positive and negative affect schedule (PANAS) en vergelijking van twee nederlandstalige versies. Gedrag En Gezondheid, 34(2), 61-70.
Folkman, S. (2008). The case for positive emotions in the stress process. Anxiety Stress and Coping, 21(1), 3-14. Fredrickson, B. (2001). The role of positive emotions in positive psychology - the broaden-and-build theory of
positive emotions. American Psychologist, 56(3), 218-226.
Hirsch, J. K., Floyd, A. R., & Duberstein, P. R. (2012). Perceived health in lung cancer patients: The role of positive and negative affect. Quality of Life Research, 21(2), 187-194.
Hou, W. K., Law, C. C., & Fu, Y. T. (2010). Does change in positive affect mediate and/or moderate the impact of symptom distress on psychological adjustment after cancer diagnosis? A prospective analysis. Psychology & Health, 25(4), 417-431.
Hou, W. K., Law, C. C., Yin, J., & Fu, Y. T. (2010). Resource loss, resource gain, and psychological resilience and dysfunction following cancer diagnosis: A growth mixture modeling approach. Health Psychology, 29(5), 484-495.
Janse, M., Fleer, J., Smink, A., Sprangers, M. A., & Ranchor, A. V. (2015). Which goal adjustment strategies do cancer patients use? A longitudinal study. Psycho‐Oncology, 25(3), 332-338.
Janse, M., Sprangers, M. A., Ranchor, A. V., & Fleer, J. (2015). Long-term effects of goal disturbance and adjustment on well-being in cancer patients. Quality of Life Research, 25(4), 1017-1027.
Janse, M., Ranchor, A. V., Smink, A., Sprangers, M. A. G., & Fleer, J. (2015). Changes in cancer patients' personal goals in the first 6 months after diagnosis: The role of illness variables. Supportive Care in Cancer: Official Journal of the Multinational Association of Supportive Care in Cancer, 23(7), 1893-1900. Jung, T., & Wickrama, K. (2008). An introduction to latent class growth analysis and growth mixture modeling.
