Psychological Factors in Chronic Edema: A Case

Abstract

Background:

To examine psychological health and perceived social support in patients with chronic edema (CE).

Methods and Results:

A random sample of 107 patients with CE was group matched for age and gender with 102 community controls. The cases had swelling of the arm (38%) and leg (61%) and midline swelling (14%). Cases were more likely to be single or divorced/separated (p = 0.041) and have reduced mobility (p < 0.001). They had significantly poorer overall health related quality of life (HRQoL), in particular physical functioning (p = 0.003); role physical (p < 0.001); general health (p = 0.026); vitality (p = 0.015); social function (p = 0.007); and role emotional (p = 0.041). EQ5-D health index scores were significantly reduced in cases by 13 points (p = <0.001, 95% confidence interval 5.8, 21.6). Cases had similar sized social networks to the control group (5.8 vs. 6.6 p = 0.49), but had lower total perceived social support scores (67.8 vs. 76.1 p = 0.018). CE patients used significantly fewer coping strategies (COPE scale) than controls with regard to the following: active coping (p = 0.024); planning (p < 0.001); and use of instrumental support (p = 0.006). Significantly higher levels of coping were used in restraint (p = 0.031), positive reinterpretation and growth (p < 0.001); acceptance (<0.001); denial (p < 0.001); mental disengagement (p < 0.001); behavioral disengagement (<0.001); substance abuse (p = 0.010); and humor (p < 0.001).

Conclusions:

Patients with CE have poorer health and greater impact on many aspects of HRQoL. Perceived social support is reduced. Deficits in social function combined with perceived reduction in support, and reduced mental health and emotional scores indicate the risk of psychological issues. Systems of care should offer an environment to address these issues.

Introduction

Chronic edema: an overview

The term chronic edema (CE) is now used interchangeably with the term Lymphedema to describe swelling present for more than 3 months and arising from the accumulation of fluid and other tissue elements that would otherwise drain through the lymphatic system.¹ It can be a grossly disfiguring condition, usually affecting the limb, which causes discomfort, functional disability, and symptoms such as pain. It can be complicated by recurrent infections often requiring hospital admission. Untreated, limbs can become huge, and the term elephantiasis illuminates their appearance. Recent research has shown that it affects many different patient populations and people are found throughout the health systems.^1,2 The cause of CE is often complex and includes the following: damage to lymphatics during cancer treatment, primary lymphedema (due to problems in the development of the lymphatic system), damage to the venous system (phlebo-lymphoedema), and obesity.^3–6 It occurs in those with chronic diseases affecting mobility such as spina bifida, multiple sclerosis, and rheumatoid arthritis, and in cardiovascular diseases.^2,7–9 The prevalence is highest in the elderly with risks factors including obesity and reduced mobility, both of which are continuing to increase in western health care populations.^1,2 Evidence from studies in cancer populations with CE indicates that the impact on health-related quality of life is high.^10–14 Other psychosocial issues have received scant attention in patients with primary lymphedema and within the heterogeneous patients affected with CE.¹⁵

The emphasis of treatment is on helping the patient achieve as much self-sufficiency and independence as possible, given the chronic nature of their condition, and to reduce the incidence of comorbidity and the need for intensive therapy and hospitalization.¹⁶ Treatment may involve an intensive phase (decongestive lymphatic therapy) that reduces the edema. Methods include appropriate skin care, manual lymphatic drainage, multilayer lymphedema bandaging, and isotonic exercise. The second phase of maintenance therapy may include self-bandaging, compression hosiery, exercise regimens, skin care, and self-administered massage. Surgery and liposuction are used in selective cases, but all require an element of patient self-management.¹⁷ Drug therapy options are currently of limited use except in selective cases.¹⁸

Psychosocial issues and chronic edema

There is increasing recognition of the role that psychological status plays in the development and outcomes of chronic disease, but little understanding of its importance in CE. Much of the quality of life research has been undertaken in patients following cancer, with a predominant focus on breast cancer-related lymphedema.^11,19,20 This has highlighted the significant impact on emotional state, social isolation, reduced functional status, and symptoms such as pain. Other issues such as the effects on body image and the ability to connect within the social environment and local culture have also been highlighted.^21,22 All these factors may affect long-term psychological health.

Reliance on data drawn from specific patient groups fails to identify the impact in the wider range of patients. It is essential to understand the clinical, psychological, and economic burden to negotiate for improved funding for this patient group in an increasingly challenged health care system. This study was designed to address this and, by including a control group, compare the additional burden of the disease.

Changing psychosocial burden

Patients with CE often have complex comorbidities with the potential to develop significant physical disability. As illness complexity increases, so may the psychological impact with the risk of problems such as depression, anxiety, and social isolation. Treatment has followed a traditional medical model that focuses on reducing tissue edema and stabilizing skin changes. Very little attention has been placed on identifying and treating psychological problems or understanding the daily challenges faced by patients, families, and carers.^23–25 There has been some attempt to demedicalize management with the development of patient support groups.²⁶

Coping, social support, and health outcomes

The link between coping, social support, and health outcomes is well established.^27,28 Epidemiological studies for decades have shown that patients with cardiovascular disease show (although not always consistently) increased mortality in those who are socially isolated and decreased in those with good levels of support.²⁹ There are a number of studies reporting an association between low social support and high levels of stress and negative health outcomes.^30–32

Access to social support has been found to be significantly associated with positive rates of recovery in a range of conditions such as myocardial infarction.^28,33,34 Burn victims reported higher self-esteem and general satisfaction if they were receiving good social support.³⁵ A study of patients with hypertension found that social support was linked to better blood pressure control and maintenance,³⁶ and patients with kidney disease were found to have higher morale if they were within a supportive, cohesive family.³⁷ Finally, a study of 110 patients with chronic obstructive airways disease³⁸ identified that social support was linked to both the functional improvement of patients as well as survival.

Coping and social support in chronic edema

There is a paucity of literature on coping and the social support needs of patients with CE, and much of what there is comes from the breast cancer literature.^10,11,23 What is apparent, however, is that the psychosocial effects of such a chronic, outwardly disfiguring and restricting condition can be profound.

Studies have shown that women who develop the condition following breast cancer treatment exhibit higher levels of psychological, social, sexual, and functional morbidity than women who do not.³⁹ Women who have poor social support, pain, CE in the dominant hand, and/or a passive and avoidant coping style report the highest level of disability.³⁹ The literature points to the importance of widening our understanding of the psychosocial impact of CE by evaluating the place of coping and perceived social support.

CE and health-related quality of life

Health-related quality of life is a concept developed in an attempt to capture the impact of health and illness on a patient's physical and mental well-being, as well as looking at the ability of the patient to function socially.⁴⁰ The emphasis is on the patient's own perception of their health and how illness impacts on all aspects of daily living, and the relationship of the individual within their social environment.⁴¹

The important point is that the severity of a condition is no longer defined solely according to clinical criteria. Assessment of health-related quality of life, as a subjective experience of the patient, is a key consideration and will vary across age groups. For example, the impact of the condition on employment, job prospects, and relationships will perhaps be important to younger patients, while mobility and concurrent illness might have a greater influence on how older patients perceive their quality of life.

In 2002, the King's Fund awarded a grant to develop and evaluate a model of care for patients with all forms of CE. The aim of this project was to develop, implement, and evaluate health service and patient outcomes using an appropriate model of care within a London-based primary care trust. The work included undertaking a psychosocial case–control study. The results are now reported.

The study set out to examine the following hypotheses:

• Patients with CE would have generally poorer psychosocial status, higher levels of disability, and reduced health-related quality of life compared to group-matched controls drawn from representative general practitioner lists

• Patients with CE would have small social networks and lower perceived social support

• Patients would use passive/avoidant coping styles, and mental and behavioral disengagement strategies compared to controls

Study aim

To investigate the association between coping styles, levels of social support, and health-related quality of life in patients with CE compared to a community-based control population.

Methods

Study design

Case identification and controls

The general methods for the identification of patients in this study are provided in a separate publication.⁴² In summary, all patients with CE were identified through primary and secondary health services and offered treatment for their CE within a new service in South West London (United Kingdom). Following the case ascertainment, 312 patients were identified and from this, a random sample of patients was drawn, stratified by gender and site of swelling. Cases were group matched to control subjects drawn from six general practitioner (GP) age-sex registers within the catchment area. Group matching occurred for gender and age. Medical notes of all controls were screened at entry for CE and if a positive history was given, they were excluded and replaced. Controls were contacted by a letter from their GP inviting them to attend the interview. Those who did not wish to take part, failed to attend the appointment, or when screened were found to have CE at the interview were replaced by new controls. The format of the interviews was identical for both cases and controls and they were undertaken by the same researcher. The analysis was undertaken by group matching rather than on an individual basis. This allowed for the slight difference in the number of cases (107) and controls (102).

Patients were offered a choice of venue for the interview, including their home or a health facility. The data collection tools included the following domains: (i). demographics, (ii) clinical details, site of CE, complications, and mobility status, and (iii) psychosocial questionnaires (EQ-5D, short form 36 [SF36], Medical Outcomes Study [MOS] social support survey, and COPE scale). Where standard questionnaires were included, the instructions for completion were followed wherever possible. Patients and controls who could complete the questionnaires, did so. Those who were unable to do so due to poor vision, were assisted in the process.

The study was supported by a steering group of relevant experts, including patients and carers. The data collection tools were piloted and amended with five patients before adaption and use. Ethical approval was obtained from the local Independent Ethics Committee, and the study was performed in accordance with the guidelines of the Declaration of Helsinki⁴³ and the principles of Good Clinical Practice ICH E6 and EN540 standards.⁴⁴

Patient recruitment and inclusion criteria

Patients were sequentially recruited to the study over a 15-month period following their identification and referral for treatment. The following inclusion and exclusion criteria were used:

Inclusion

• Patients with confirmed CE of >3 months affecting the limbs and/or the trunk irrespective of the underlying cause

• Patients able to understand the study and give written informed consent

Exclusion

• Patients with advanced or terminal cancer

• Patients who were unable to speak English or complete the questionnaires

• Patients unable to give written informed consent

Data collection tools

A range of validated health related quality of life (HRQoL) and psychosocial tools were used to fulfill the objectives of the study.

The MOS SF36

The SF36 was chosen as a generic tool to assess health-related quality of life because it has been used in other Lymphedema populations.⁴⁵ This is a well-validated tool that provides estimates of eight health concepts, these being; physical functioning (PF), role functioning-physical (PF), bodily pain, general health (GH), vitality (VI), social functioning, role functioning-emotional, and mental health. The responses to the 36 questions are transformed into the concepts to provide a scale from 0 (worst possible health) to 100 (best possible health).

The Euroqol (EQ-5D)

The Euroqol (EQ-5D) was used to measure health status; the tool consists of five questions in the domains of mobility, self-care, usual activities, pain, and anxiety.⁴⁶ The patients' health state is determined from the responses to these questions and fitted into an equation developed from a regression model. It produces scores from −0.59 to 1.00, the lowest score being the worst possible health state and 1.00 being the best. Death is valued as zero within this scale. In addition, participants are asked to rate their health on a scale from zero to 100.

The MOS social support survey

Despite the plethora of social support tools that have been developed, few have been shown to have robust psychometric properties. This situation is compounded by the lack of clear conceptual agreement over what constitutes social support.⁴⁷ The MOS has been developed in an attempt to overcome these issues. The tool is designed to represent the multiple dimensions of functional support: emotional/informational, tangible, affectionate, and positive social interaction.

The MOS is divided into two sections. The first section requires the patient to number how many close friends and relatives they feel at ease with and can talk to about what is on their mind. This is defined as the social network measure. The second section involves 19 statements concerning different aspects of social support. Patients are required to make a response, nearest to their current experience, on a five-point Likert scale (None of the time–all the time). Each of the subscales in the tool is scored individually. Scores are ordinally arranged with possible scores for the total 19 questions ranging from 19 to 95.

The COPE scale

The COPE scale was used to assess coping.⁴⁸ It was chosen because it is a theoretically devised and validated scale that seeks to examine the multidimensional nature of coping and has been used in different chronic illness populations. The tool was designed to overcome the simplistic view that coping could be divided into problem focused and emotion focused coping. The tool has been validated as a situational (a real-life event) as well as a dispositional tool (a generalized response to stressful events). This is particularly important in this study in which a control population was being examined.

The COPE scale consists of 13 separate domains, including five subdivided as problem focused (active coping; planning; seeking instrumental support; suppression of competing activities; and restraint coping) and five as emotion focused (positive reinterpretation and growth; acceptance; focus on venting emotion; and denial). Further domains are mental disengagement; behavioral disengagement; alcohol; seeking emotional support; humor; and turning to religion.

The patients with CE completed a situational-based COPE scale where they were asked to consider how they coped with having the condition. The control population completed a dispositional version of the scale. This attempts to determine how people respond to difficult or stressful events in their lives.

The questionnaire was self-administered in the study. Separate scores for each scale were calculated by adding the four scores from the items within each scale. The scores for each scale ranged from 4 to 16. The scores for each scale indicate the extent to which the type of coping is being used. Scores from the situational COPE were compared with the dispositional data from the control group to determine differences in the two populations.

Data analysis

Categorical data were analyzed using logistic regression analysis with the dependent variable being the participants' CE status (case or control). Results were reported as both percentages of the total in each category together with odds ratios and 95% confidence intervals. The level of significance was derived from the logistic regression result.

Continuous data were analyzed comparing the two groups (cases and controls). Results were presented as means (standard deviation [SD]) together with mean difference and 95% confidence intervals. The level of statistical significance was derived from the t-test analysis.

Bias

Attempts to reduce bias within the study were undertaken. This included drawing a representative sample from the wider group for gender and distribution of the site of swelling. Controls were drawn from the same geographical area and GP registers and matched for age and gender to the cases. It is recognized that people with early CE will not be identified through this method as they may not have yet presented to health professionals; therefore the patients will be more severe than the total population affected.

Dealing with missing data

Due to the frailty of some of the patients, not all questionnaires were completed. This is reflected in the presentation of results in which the number of cases and controls completing the questions are identified within the tables. Health-related quality of life results were adjusted for age and gender,

Results

From the main case ascertainment through health services, 312 patients who fulfilled the criteria for CE were identified. To reach the required sample of 107 cases, 120 people were required to be screened. The results for the cases and controls are presented in Table 1. The main reason for nonselection for cases was failure to attend appointments or an unwillingness or inability to participate in the study. Cases were replaced according to the stratification until the sample was completely recruited. A similar procedure was used for controls in addition to screening for previous CE. The reasons for noninclusion were similar to the cases. A total of 128 controls were screened to achieve the control sample.

Table 1.

Comparison of Demographic and Social Factors

	Controls (N = 102), n (%)	Cases (N = 107), n (%)	OR	95% CI	p
Gender
Men	19 (19)	19 (17)	1.00	0.52–2.14	0.87
Women	83 (81)	88 (82)	1.06	0.52–2.14	0.87
Age
<60	27 (26)	27 (25)	1.00
60–69	24 (24)	27 (25)	1.13	0.52–2.42
70–79	24 (24)	21 (20)	0.88	0.40–1.93	0.88
80+	27 (26)	32 (30)	1.19	0.57–2.48
Marital status
Widowed	30 (30)	26 (24)	1.00
Single	11 (11)	20 (19)	2.09	0.85–5.18	0.041
Married	52 (51)	42 (39)	0.93	0.48–1.81
Divorced/separated	8 (8)	19 (18)	2.74	1.03–7.29
Living status
Lives alone	37 (36)	42 (39)	1.00
With spouse	51 (50)	45 (42)	0.78	0.43–1.41	0.64
Relative/companion	10 (10)	15 (14)	1.32	0.53–3.30
Other	4 (4)	5 (5)	1.10	0.28–4.41
Mobility status (Euroqol)
No problems walking	64 (64)	38 (36)	1.00
Problems walking/confined to bed	36 (36)	68 (64)	3.18	1.80–5.62	<0.001

Values expressed as OR and 95% CIs generated from logistic regression analysis.

OR, odds ratio; CI, confidence interval.

Within the sample of cases (n = 107), 88 (82%) were women. The mean (SD) age was 72.9 years in men and 68.6 years in women. Within the sample, 77 (75%) were older than 70 years and 27 (25%) >80 years. Only 27 (25%) were under the age of 60. As expected, given the age of the sample, most people were retired (78%) with an additional 4.7% not working due to illness.

CE affected the leg in nearly two-thirds, 63 (59%), with 41 (38%) having arm swelling and a further 15 (14%) having both midline swelling in addition to the limbs being affected (Table 2). Cancer or its treatment was the cause of CE in 55 (52%), in whom 35 (69%) had controlled cancer with only 1 (2%) suffering metastatic disease. The swelling was of long duration with 50 (47%) suffering for more than 10 years and a further 54 (51%) for between 1 and 10 years. Only 2 (2%) had swelling for a short period (3–11 months). Cellulitis episodes occurred in 22 (21%) cases, of whom 5 (5%) required hospital admission.

Table 2.

Clinical Details of the Swelling

	Cases (N = 107), n (%)
Site of swelling
Arm(s)	41 (38)
Leg (s)	63 (59)
Arm(s) + leg(s)	3 (3)
Limb plus midline	15 (14)
Cancer cause
No	55 (52)
Yes	51 (48)
Cancer status
Controlled	35 (69)
Inactive	14 (27)
Metastatic	1 (2)
Not known	1 (2)
Swelling duration
3–11 months	2 (2)
1–5 years	40 (38)
6–9 years	14 (13)
>10 years	50 (47)
Infection over past year
None	85 (79)
One	14 (13)
Two	3 (3)
3–5	2 (2)
6–10	2 (2)
>10	1 (1)
Admission for infection over past year
None	102 (96)
One	4 (4)
Two	1 (1)

There were significant differences in the marital status of cases and controls (p = 0.041). More cases were single (19), divorced, or separated (18) compared to controls (11 vs. 8) (p = 0.041). No statistically significant difference was found in the living status between cases and controls although more cases were likely to live with a relative or companion (15 vs. 10) (p = 0.64). Cases had significant deficits in mobility with many confined to a chair or requiring a walking aid to achieve any level of mobility. The mobility status of the control group was significantly better (68 vs. 36) (p < 0.001).

Short form 36

Results from the SF36 reveal significant deficits in the HRQoL scores in cases compared with the control group (Table 3). In particular, deficits were seen in physical functioning (p = 0.003), role physical (p < 0.001), general health (p = 0.026), vitality (p = 0.015), social function (p = 0.007), and role emotional (p = 0.08). However, all other domains were reduced and approaching a standard level of significance.

Table 3.

Comparison of the Short Form 36 Between Cases with Chronic Edema and Controls

	Control		Case
	N	Mean (SD)	N	Mean (SD)	Difference	95% CI	t (df)	p
Physical functioning	102	67.6 (43.6)	107	48.9 (46.8)	18.7	6.4–31.1	2.99 (207)	0.003
Role physical	101	71.2 (29.1)	107	45.9 (33.3)	25.3	16.7–33.9	5.82 (206)	<0.001
Bodily pain	101	70.2 (29.9)	107	64.0 (30.9)	6.2	−2.1 to 14.5	1.47 (206)	0.14
General health	101	67.1 (22.9)	107	59.5 (26.2)	7.7	1.0–14.4	2.25 (206)	0.026
Vitality	101	59.4 (21.5)	107	51.5 (24.2)	7.8	1.5–14.1	2.46 (206)	0.015
Social functioning	101	78.7 (31.5)	107	66.7 (32.5)	12.1	3.3–20.8	2.71 (206)	0.007
Role emotional	102	78.4 (39.7)	107	66.4 (44.7)	12.1	0.5–23.6	2.06 (207)	0.041
Mental health	101	78.1 (19.7)	107	73.3 (19.2)	4.8	−0.5 to 10.1	1.78 (206)	0.077
Euroqol	99	79.3 (23.5)	103	65.6 (32.7)	13.7	5.8–21.6	3.40 (200)	<0.001
Visual analog scale	100	76.5 (15.5)	107	64.1 (18.9)	12.4	7.6–17.1	5.14 (205)	<0.001

SD, standard deviation.

EQ-5D

EQ-5D health index scores were significantly reduced in cases by 13 points compared with controls (79 vs. 66 p < 0.001). Results from the visual analog scale (VAS) show a similar difference, 76 versus 64 (<0.001). The mobility questions confirmed the significant impact on mobility with 68 (64%) of cases stating they had problems with walking or were confined to a wheelchair or bed, compared to only 36 (36%) of the controls (p < 0.001).

MOS social support survey

The number of people within the social networks of cases and controls was similar (5.8 vs. 6.6) (p = 0.49). The total scores for cases (67.8) were significantly lower than for the controls (76.1), suggesting an overall lower perceived social support (p = 0.018). Table 4 shows that cases reported reduced social support for all the scales, except for emotional/informational support (p = 0.09) and positive interaction (p = 0.07), both of which were approaching a standard level of statistical significance.

Table 4.

Differences in the Mean Medical Outcomes Study Social Support Survey Scores (Case–Control)

	Control		Cases		Case–Control		t	df	p
	N	Mean (SD)	N	Mean (SD)	Mean difference (SEM)	95% Confidence interval	t	df	p
Emotional/informational	96	73.5 (28.8)	103	66.3 (31.5)	7.2 (4.3)	−1.3 to 15.7	1.68	197	0.09
Tangible support	100	81.4 (24.6)	100	70.3 (29.1)	11.1 (3.8)	3.5–18.6	2.9	198	0.004
Positive interaction	98	73.6 (28.8)	105	65.6 (32.4)	8.0 (4.3)	−0.5 to 16.5	1.85	201	0.07
Affectionate support	99	83.1 (27.7)	102	72.4 (32.5)	10.7 (4.3)	2.3–19.1	2.51	199	0.013
Total score	93	76.1 (25.4)	95	67.8 (28.6)	8.3 (4.0)	0.5–16.1	2.10	186	0.018
Number of people	98	6.6 (4.5)	106	5.8 (9.8)	0.7 (1.1)	−1.4 to 2.9	0.68	202	0.49

Values given are the mean difference (SEM) and 95% CIs, t- and p-value.

SEM, standard error of the mean.

COPE scale

Results from the COPE scale indicate that the patients used less problem- and emotion- focused coping strategies than the control group (Table 5). Cases used significantly fewer coping strategies than controls with regard to the following: active coping (p = 0.024); planning (p < 0.001); and use of instrumental support (p = 0.006). Significantly higher levels of coping were used by cases for the following domains: restraint (p = 0.031); positive reinterpretation and growth (p < 0.001); acceptance (p < 0.001); denial (p < 0.001); mental disengagement (p < 0.001); behavioral disengagement (p < 0.001): substance abuse (p = 0.010); and humour (p < 0.001). Religious coping strategies were similar between groups (p = 0.56).

Table 5.

Differences in COPE Scores (Case–Control)

	Control		Cases		Case–Control		t	df	p
	N	Mean (SD)	N	Mean (SD)	Mean difference (SEM)	95% CI	t	df	p
Problem focused
Active coping	100	10.8 (2.7)	103	9.8 (3.2)	1.0 (0.4)	0.1 to 1.8	2.28	201	0.024
Planning	100	10.3 (3.0)	102	8.7 (3.5)	1.6 (0.5)	0.7 to 2.5	3.58	200	<0.001
Use of instrumental social support	102	8.6 (2.9)	105	7.5 (2.8)	1.1 (0.4)	0.3 to 1.9	2.77	205	0.006
Suppression of competing activities	101	6.5 (3.2)	102	6.9 (3.0)	−0.4 (0.4)	−1.2 to 0.46	0.92	201	0.36
Restraint	101	7.0 (3.3)	100	7.9 (2.9)	−1.0 (0.4)	−1.8 to −0.1	2.18	199	0.031
Positive reinterpretation and growth	102	8.9 (3.6)	102	10.7 (3.5)	−1.8 (0.5)	−2.8 to −0.8	3.66	202	<0.001
Acceptance	99	8.7 (4.3)	104	12.2 (3.1)	−3.5 (0.5)	−4.4 to −2.4	6.59	201	<0.001
Focus on venting of emotions	102	7.2 (3.1)	103	6.8 (3.0)	0.4 (0.4)	−0.4 to 1.3	0.99	203	0.32
Denial	102	4.6 (1.4)	105	6.0 (3.0)	−1.4 (0.3)	−2.1 to −0.8	4.43	205	<0.001
Mental disengagement	102	6.3 (2.2)	104	7.8 (2.7)	−1.4 (0.3)	−2.1 to −0.7	4.13	204	<0.001
Behavioral disengagement	101	4.6 (1.2)	103	5.7 (2.2)	−1.1 (0.2)	−1.6 to −0.6	4.55	202	<0.001
Substance use	102	4.2 (0.7)	104	4.7 (1.8)	−0.5 (0.2)	−0.9 to −0.1	2.60	204	0.010
Use of emotional social support	101	6.6 (3.4)	105	7.1 (2.9)	−0.5 (0.4)	−1.4 to 0.4	1.16	204	0.25
Humor	102	5.7 (2.6)	104	7.8 (3.3)	−2.0 (0.4)	−2.8 to −1.2	4.89	204	<0.001
Religious coping	101	8.4 (5.0)	102	8.8 (4.4)	−0.4 (0.7)	−1.7 to 0.9	0.58	201	0.56

Values given are the mean difference (SEM) and 95% confidence intervals, t- and p-values.

Discussion

The results from this study would indicate that CE is associated with significant psychosocial morbidity. The demographic data from the cases confirmed that this was a predominantly elderly population who had been suffering with CE for a long period with less than half having CE as a consequence of cancer or its treatment. The proportion of women was high and there was evidence of significant impact on mobility. General health status was poor with a 13-point difference in EQ-5D scores compared to controls. This result was confirmed by the VAS scores. Elderly populations will experience a range of chronic diseases; however, the control group reports an improved health profile, suggesting that the patients with CE have a complex array of medical conditions resulting in swelling that is impacting many aspects of life.

The study found substantial deficits in HRQoL and confirmed the first study hypothesis that patients with CE would have generally poorer psychosocial status, higher levels of disability, and reduced health-related quality of life compared to GP age/sex group matched controls drawn from representative general practitioner lists. The controls were drawn from the same geographical population at the same time. The differences observed therefore cannot be explained by temporal or regional differences between the two groups. Incorporation of a control group into this study has strengthened the understanding of the magnitude of effect of CE compared with people of similar age and gender. Results from the EQ-5D scores indicated the cases had significant health deficits. In a study to define utility scores in 277 patients with different causes of lymphedema using the EQ-5D scale, the lowest scores occurred for those with lower limb CE secondary to cancer, with the greatest in those with severe swelling and a high body mass index (BMI). The EQ-5D scores indicate a worse health status than in other published studies that report a mean score in the range of 80 compared to 66 in this study.⁴⁹

The second study hypothesis was that patients with CE would have small social networks and lower perceived social support. This has been partially supported with cases indicating low levels of social support; however, the size of the social networks was similar to the control group. Patients with poor psychosocial support available and low levels of information have been found to have high levels of anxiety and depression.⁵⁰ In this study, cases had significantly lower levels of perceived support than the controls, suggesting that this may be an influencing factor. High levels of depression have consistently been associated with passive patient behavior such as nonadherence to therapy.⁵¹

The results from the COPE questionnaire support the third study hypothesis that patients with CE would use more passive/avoidant coping styles and mental and behavioral disengagement strategies compared to controls. The overall pattern of results suggests that the patients with CE who completed a questionnaire based on their current experience, used fewer coping strategies than the control group who completed the dispositional version of the tool. It is not possible to determine how much the perceived differences can be explained by the use of a situational questionnaire in patients who are experiencing a real situation compared to those who are asked to make a judgment about how they generally cope with stressful situations. The control group may be overpredicting the use of coping strategies that they believe they would use. In addition, the cases were all at different points in their illness experience. Evidence from the coping research has shown that patients with chronic illness tend to use more emotion-focused strategies than problem focused.⁵² Patients may have undergone considerable adaptation that is influencing the pattern of coping strategies they use in this study.

Little research has been undertaken to determine how patients cope with their CE with some evidence from the breast cancer population of a complex pattern of usage with men showing lower coping scores.⁵³ Those with melanoma and CE affecting the leg had worse quality of life and lower coping scales compared to those in whom the tumor was on the arm.²⁴ In a qualitative study of leg ulcer patients, Walshe described how patients used strategies that sought to normalize their condition to cope with the relentless demands of their condition and treatment.⁵⁴ Four main strategies were identified: coping by comparison, coping by feeling healthy, coping by altered expectation, and coping by remaining positive.

The outcome of individual coping strategies on symptoms associated with CE is likely to be complex, to vary over time, and to be influenced by the disease trajectory. Suppression of activities that may increase the prominence of symptoms such as pain may be an important coping strategy that allows patients to live with their CE with less intrusion of symptoms and potentially with an improved quality of life; however, these may not be beneficial to overall control of edema. Strategies such as exercise, which may be viewed as important by professionals, may be seen differently by patients who see no perceived benefit, but do note that such activities increase their symptoms. This raises issues concerning how health beliefs influence the choice of coping strategy. It reinforces the issue that coping must be viewed flexibly and the relevance of individual coping strategies may be influenced by many factors, not least the stage and severity of the patient's illness.

Patients appear to be using a similar pattern of coping to those found in other chronic illness populations, strategies that may be considered more maladaptive such as use of alcohol and drugs, and continuous periods of mental disengagement and denial were used in moderation.³²

The results would suggest that the patients are socially isolated. It is not possible to determine whether the social isolation is a result of reduced social support or a consequence of the condition, which has been shown in other chronic illnesses. It could also be speculated that the patients are voluntarily withdrawing from social contact because they are aware of people's reaction to their condition. The debilitating long-term effects of symptoms such as pain may also be influencing their desire for social interaction. Problems of mobility are severe in this patient group and may influence their ability to socialize if they wished to.

The low levels of seeking instrumental and emotional support are evidenced despite the preservation of the size of their social networks. This indicates a more subtle effect on how people living with CE perceive support. The size of the social network may indicate that they do not feel supported, which is in turn reducing their overall quality of life status and increasing the perceptions of social isolation. Low levels of perceived support are likely to affect adherence to treatment and self-efficacy and influence the effectiveness of the coping strategies that are adopted. These psychological factors, in turn, may influence clinical outcomes such as control of swelling and rates of complications such as cellulitis. For many, these issues may trap them into a spiral of clinical and psychosocial deterioration.

The role of social support in control of CE has received little research attention. In patient populations with wounds, many of whom have concurrent CE, there is increasing evidence that social support may affect health outcomes. A number of studies in leg and foot ulceration report low levels of perceived support.⁵⁵ In addition, there is some evidence of an increased rate of healing in those with high levels of support, while those with high levels of pain and depression had reduced healing even when adjusted for age and underlying etiology.^55,56

Of the 107 patients, a number were unable to complete the questionnaires due to a range of medical and psychological issues. It has long been recognized that these patients may be suffering from more extensive problems, but due to their condition, they are unable to verbalize this. There remains a major research challenge to capture the impact of their suffering. This study was cross sectional in design, with patients being at different stages of their illness diathesis. Clearly, the psychological status may vary over time, as will their social interactions. Tools such as the SF36 are generic and therefore may not identify the more subtle quality-of-life issues associated with the specific condition. These tools also do not allow us to understand what the patients' own expectations are about their condition, what would improve their sense of well-being, and which aspects of HRQoL they would like to see improved.

Current research has focused on the evaluation of HRQoL and attempts to determine changes with treatment. Clearly, there is a need to understand in more depth the aspirations of patients to ensure that they can function to their highest ability. This study would support that clinicians need to have greater awareness of psychological health problems in patients with CE and services need to reflect the issues identified, with access to psychological support to those showing distress.

Limitations of the study

This study has a number of limitations. The sample size, although reflective of the larger population, is relatively small. The patients recruited are likely to be complex in their clinical presentation. However, despite this fact, there is evidence that they are typical of the types seen in current CE services in the United Kingdom.^1,2 Many have had their condition for a long duration and have not had an appropriate diagnosis or treatment leading to deterioration and morbidity.¹ As there are no agreed coding systems or requirements to screen for CE in primary care in the United Kingdom, it is likely that many with mild CE will not be identified and only become apparent when complications occur. Cases and controls were identified in the health care system and it is acknowledged that not everyone in the United Kingdom is registered with a GP and transitory populations are not well represented in this type of research. The study used well-validated generic HRQoL tools; however, more recently disease-specific tools have been developed that may be more sensitive to change and should be incorporated into future research.⁵⁸⁵⁹ Despite this, the tools adopted for the study have been able to identify significant differences between groups and importantly with other disease states. The questionnaires took <1 hour to complete. While there were some missing data, the completion rate was generally in excess of 90%. In addition, there were some data that were not collected. The measurement of BMI was not undertaken in the controls due to logistic issues, and this may have provided information on how this may have impacted the patients with CE.

The data in this study have been used to compare cases with CE with controls who do not. There is a temptation to look at further analysis. In particular, one may wish to examine subgroups of the total to identify other factors such as the type of limb involvement and the presence of cancer treatment as a cause of deficits in social support and coping. Due to small numbers these were not undertaken and would require a larger sample size and a rigorous protocol to address these issues.

Conclusion

This study has highlighted the psychological impact that CE has on patients. There is a clear need to examine this in more detail, particularly with respect to the differences experienced at varying stages of the illness trajectory and in the different manifestations of CE related to the array of associated medical conditions. The low levels of perceived support and increased levels of social isolation support the concept of services that allow for an integrated approach, which includes psychological support and offers patient interaction, a relief from isolation, and an opportunity to share common experiences.

Footnotes

Acknowledgment

This study was undertaken as part of a King's Fund grant.

Author Disclosure Statement

No competing financial interests exist.

References

Moffatt

, Franks

, Doherty

, Williams

, Badger

, Jeffs

, Bosanquet

, Mortimer

. Lymphoedema: An underestimated health problem. Q J Med, 2003; 96:731–738.

Moffatt

, Keeley

, Franks

, Rich

, Pinnington

. Chronic oedema: A prevalent health care problem for UK health services. Int Wound J, 2016; [Epub ahead of print]; DOI: 10.1111/iwj.12694

Armer

, Stewart

, Shook

. 30-Month post-breast cancer treatment lymphoedema. J Lymphoedema, 2009; 4:14–18.

Smeltzer

, Stickler

, Schirger

. Primary lymphedema in children and adolescents: A follow-up study and review. Pediatrics, 1985; 76:206–218.

Bull

, Gane

, Evans

, Joseph

, Mortimer

. Abnormal lymph drainage in patients with chronic venous leg ulcers. J Am Acad Dermatol, 1993; 28:585–590.

Boza

, Trindade

, Peruzzo

, Sachett

, Rech

, Cestari

. Skin manifestations of obesity: A comparative study. J Eur Acad Dermatol Venereol, 2012; 26:1220–1223.

Garcia

, Dicianno

. The frequency of lymphedema in an adult spina bifida population. Am J Phys Med Rehabil, 2011; 90:89–96.

Keeley

, Pinnington

, Riches

, Phillips

, Fakis

, Vaithianathar

, Jones

, Moffatt

The prevalence of chronic lower limb oedema (CLLO) in multiple sclerosis (MS). Abstract World Congress of Lymphology, San Francisco; 2015.

Keast

, Despatis

, Allen

, Brassard

. Chronic edema/lymphedema: Under-recognised and under-treated. Int Wound J, 2015; 12:328–333.

10.

Cemal

, Jewell

, Albornoz CR Pusic

, Mehrara

. Systematic review of quality of life and patient reported outcomes in patients with oncologic related lower extremity lymphedema. Lymphat Res Biol, 2013; 11:14–19.

11.

Pusic

, Cemal

, Albornoz

, Klassen

, Cano

, Sulimanoff

, Hernandez

, Massey

, Cordeiro

, Morrow

, Mehrara

. Quality of life among breast cancer patients with lymphedema: A systematic review of patient-reported outcome instruments and outcomes. J Cancer Surviv, 2013; 7:83–92.

12.

Taghian

, Miller

, Jammallo

, O'Toole

, Skolny

. Lymphedema following breast cancer treatment and impact on quality of life: A review. Crit Rev Oncol Hematol, 2014; 92:227–234.

13.

Dunberger

, Lindquist

, Waldenstrom

A-C

, Nyberg

, Steineck

, Avall-Lundqvist

. Lower limb lymphedema in gynecological cancer survivors—effect on daily life functioning. Support Care Cancer, 2013; 21:3063–3070.

14.

, Kang

. Psychosocial impact of living with cancer-related lymphedema. Semin Oncol Nurs, 2013; 29:50–60.

15.

Okajima

, Hirota

, Kimura

, Inagaki

, Tamai

, Iizaka

, Nakagami

, Mori

, Sugama

, Sanada

. Health-related quality of life and associated factors in patients with primary lymphedema. Jpn J Nurs Sci, 2013; 10:202–211.

16.

International Lymphoedema Framework. International Consensus: Best Practice for the Management of Lymphoedema. London, United Kingdom: Medical Education Partnership (MEP) Ltd.; 2006.

17.

Cormier

, Damstra

, Brorson

, Suami

, Chang

. Surgical Intervention: A Position Document on Surgery for Lymphoedema. United Kingdom: International Lymphoedema Framework; 2012.

18.

International Society of Lymphology. The diagnosis and treatment of peripheral lymphedema. Consensus document of the International Society of Lymphology. Lymphology, 2003; 36:84–91.

19.

Klernas

, Johnsson

, Horstmann

, Kristjanson

, Johansson

. Lymphedema quality of life inventory (LyQLI)—development and investigation of validity and reliability. Qual Life Res 2015;, 24; 427–439.

20.

Moseley

, Carati

, Piller

. A systematic review of common conservative therapies for arm lymphoedema secondary to breast cancer treatment. Ann Oncol, 2007; 18:639–646.

21.

Jager

, Doller

, Roth

. Quality-of-life and body image impairments in patients with lymphedema. Lymphology, 2006; 39:193–200.

22.

Person

, Bartholomew

, Addiss

, van den Borne Bart

. Disrupted social connectedness among Dominican women with chronic filarial lymphedema. Patient Educ Couns, 2007; 68:279–286.

23.

Jeffs

, Ream

, Shewbridge

, Cowan-Dickie

, Crawshaw

, Huit

, Wiseman

. Exploring patient perception of success and benefit in self-management of breast cancer-related arm lymphoedema. Eur J Oncol Nurs, 2016; 20:173–183.

24.

Cromwell

, Chiang

, Mungovan

, Ross

, Gershenwald

, Lee

, Royal

, Lucci

, Cormier

, Armer

, Heppner

. Is surviving enough? Coping and impact on activities of daily living among melanoma patients with lymphoedema. Eur J Cancer Care, 2015; 24:724–733.

25.

Chachaj

, Malyszczak

, Pyszel

, Lukas

, Tarkowski

, Pudelko

, Andrzejak

, Szuba

. Physical and psychological impairments of women with upper limb lymphedema following breast cancer treatment. Psychooncology, 2010; 19:299–305.

26.

Lindsay

, Hawkins

. Care study: The leg club model and the sharing of knowledge. Br J Nurs, 2003; 12:784–790.

27.

House

, Robbins

, Metzner

. The association of social relationships and activities with mortality. Prospective evidence from the Tecumseh Community Health Study. Am J Epidemiol, 1982; 116:123–140.

28.

Blazer

DG.

Social support and mortality in an elderly population. Am J Epidemiol, 1982; 115:684–694.

29.

Marmot

, Shipley

, Rose

. Inequalitites in death. Specific explanations of a general pattern. Lancet, 1984; 1:1003–1006.

30.

Haynes

, Feinleib

, Kannel

. The relationship of psychosocial factors to coronary artery disease in the Framingham Study III. Eight year incidence of coronary heart disease. Am J Epidemiol, 1980; III:37–58.

31.

Welin

, Svardsudd

, Ander-Peciva

, Tibblin

, Larsson

. Prospective study of social influences on mortality. Lancet, 1985; 2:915–918.

32.

Cohen

, Syme

. Issues in the study and application of social support. In: Cohen

, Syme

, eds. Social Support and Health. San Diego, CA: Academic Press; 1985:3–22.

33.

Berkman

, Leo-Summers

, Horwitz

. Emotional support and survival after myocardial infarction. A prospective, population-based study of the elderly. Ann Intern Med, 1992; 117:1003–1009.

34.

Montbriand

, Laing

. Alternative health care as a control strategy. J Adv Nurs, 1991; 16:325–332.

35.

Davidson

, Bowden

, Tholen

. Social support as a moderator of burn rehabilitation. Arch Phys Med Rehabil, 1979; 60:556.

36.

Finlayson

Social support networks as coping resources: Lay help and consultation patterns used by women in husband's post-infarction careers. Soc Sci Med, 1976; 10:97–103.

37.

Diamond

Social support and adaption to chronic illness: The case of maintenance haemodialysis. Res Nurs Health, 1979; 2:101–108.

38.

Grodner

, Prewitt

, Jaworski

, Myers

, Kaplan

, Ries

. The impact of social support in pulmonary rehabilitation of patients with chronic obstructive pulmonary disease. Ann Behav Med, 1996; 18:139–145.

39.

Passik

, McDonald

. Psychosocial aspects of upper extremity lymphoedema in women treated for breast carcinoma. Cancer, 1998; 83:2817–2820.

40.

Fallowfield

The Quality of Life. The Missing Dimension in Health Care. London, United Kingdom: Sonnvenir; 1990.

41.

World Health Organisation. Rehabilitation after Cardiovascular Diseases, with Special Emphasis on Developing Countries: Report of a WHO Expert Committee. WHO Technical Report Series 831. Geneva: WHO; 1993.

42.

Moffatt

, Doherty

, Franks

, Mortimer

. Community based treatment for chronic oedema–an effective service model (In Press).

43.

Declaration of Helsinki. World Medical Association. Available at: http://wma.net/publications/wma-doh-1964-2014/. (2014). Last accessed September 7, 2017.

44.

Guidelines for Good Clinical Practice. ICH Harmonised Tripartite Guidelines. Geneva, Switzerland: ICH. 1996.

45.

Ware

, Snow

, Kosinski

, Gandek

. SF-36 Health Survey: Manual and Interpretation Guide. Boston, MA: The Health Institute, New England Medical Center; 1993.

46.

The Euroqol Group. EuroQol—a new facility for the measurement of health-related quality of life. The EuroQol Group. Health Policy, 1990; 16:199–208.

47.

Sherborne

, Stewart

. The MOS social support survey. Soc Sci Med, 1991; 32:705–714.

48.

Carver

, Scheier

, Weintraub

. Assessing coping strategies: A theoretically based approach. J Person Soc Psychol, 1989; 56:267–283.

49.

Cheville AL Almoza

, Courmier

, Basford Jeffrey

. A prospective cohort study defining utilities using time trade-offs and the Euroqol-5D to assess the impact of cancer-related lymphedema. Cancer, 2010; 116:3722–3731.

50.

Tobin

, Lacey

, Meyer

, Mortimer

. The psychological morbidity of breast cancer-related arm swelling, psychological morbidity of lymphoedema. Cancer, 1993; 72:3248–3252.

51.

DiMatteo

, Lepper

, Croghan

. Depression is a risk factor for noncompliance with medical treatment: Meta-analysis of the effects of anxiety and depression on patient adherence. Arch Int Med, 2000; 160:2101–2107.

52.

Maes

, Leventhal

, de Ridder

DTD

. Coping with chronic diseases. In: Zeidner

, Endler

, eds. Handbook of Coping. Theory, Research and Applications. New York: John Wiley; 1996:221–225.

53.

Dominick

SA.

, Natarajan

, Pierce

, Madanat

, Madlensky

. The psychosocial impact of lymphedema-related distress among breast cancer survivors in the WHEL Study. Psychooncology, 2014; 23:1049–1056.

54.

Walshe

Living with a venous leg ulcer: A descriptive study of patient's experiences. J Adv Nurs, 1995; 22:1092–1100.

55.

Keeling

, Price

, Jones

, Harding

. Social support. Some pragmatic implications for health care professionals. J Adv Nurs, 1996; 23:76–81.

56.

Hyland

, Thomson

. Quality of life of leg ulcer patients: Questionnaire and preliminary findings. J Wound Care, 1994; 3:294–298.

57.

Weiss

, Daniel

. Validation of the lympedema life impact scale (LLIS): A condition-specific measurement tool for persons with lymphedema. Lymphology, 2015; 48:128–138.

Psychological Factors in Chronic Edema: A Case–Control Study

Abstract

Abstract

Background:

Methods and Results:

Conclusions:

Introduction

Chronic edema: an overview

Psychosocial issues and chronic edema

Changing psychosocial burden

Coping, social support, and health outcomes

Coping and social support in chronic edema

CE and health-related quality of life

Study aim

Methods

Study design

Case identification and controls

Patient recruitment and inclusion criteria

Inclusion

Exclusion

Data collection tools

The MOS SF36

The Euroqol (EQ-5D)

The MOS social support survey

The COPE scale

Data analysis

Bias

Dealing with missing data

Results

Short form 36

EQ-5D

MOS social support survey

COPE scale

Discussion

Limitations of the study

Conclusion

Footnotes

Acknowledgment

Author Disclosure Statement

References