| |
Factors Associated with Severe Impact of Lipodystrophy on the Quality of Life of Patients Infected with HIV-1
|
| |
| |
Clinical Infectious Diseases April 2004;38:1469-1475
Jordi Blanch,1 Araceli Rousaud,2 Esteban Martínez,3 Elisa De Lazzari,4 Ana Milinkovic,3 Josep-Maria Peri,1 José-Luís Blanco,3 Jesús Jaen,1 Victor Navarro,1 Guillem Massana,1 and Josep-Maria Gatell3
1Clinical Institute of Psychiatry and Psychology, 2Fundació Clínic per a la Recerca BiomèdicaInstitut d'Investigacions Biomèdiques Agustí Pi i Sunyer, 3Clinical Institute of Infectious Diseases and Immunology, and 4Epidemiology and Biostatistics Unit, Hospital Clínic Universitari de Barcelona, Spain
ABSTRACT
A standardized questionnaire was used to assess the impact of lipodystrophy (LD) on quality of life (QoL). Eighty-four consecutive asymptomatic human immunodeficiency virus type 1 (HIV-1)infected outpatients with clinical LD completed a modified version of the Dermatology Life Quality Index (DLQI) survey to measure the impact of body fat changes on their QoL. Body changes influenced dressing for 55 patients (65%), produced feelings of shame for 41 (49%), and disrupted sexual life for 23 (27%). There was a greater impact on the DLQI due to body changes among women, injection drug users, patients with abdominal or breast lipoaccumulation, and patients with a high number of non-LD side effects. Multivariate proportional odds model analysis showed that the severity of nonLD-associated side effects and the presence of breast lipoaccumulation were associated with impaired psychosocial functioning. Specific characteristics of patients, antiretroviral-based side effects, and breast lipoaccumulation exert a greater impact on QoL in HIV-1infected patients with LD.
Now that HIV type 1 (HIV-1) infection is considered to be a chronic illness, quality of life (QoL) has become an important outcome measurement. Few instruments specifically measure QoL in HIV-1infected patients, and none are specific for lipodystrophy (LD) in such patients. HAART can produce side effects that may impair QoL and threaten compliance; in spite of this, <3% of drug trial publications incorporate measurement of QoL as an outcome variable. Instruments that assess body image should be developed to measure QoL in HIV-1infected patients, and more attention should be paid to specific dimensions of QoL in the management of HIV-1infected patients.
Although LD may affect QoL, in a previous study, we failed to show differences in overall QoL associated with the presence or absence of the disorder. However, we noted that certain patient characteristics were associated with a greater impact on QoL. Therefore, to broaden our understanding of the impact of LD on QoL and to facilitate design of psychotherapeutic strategies to improve or preserve QoL in these patients, we analyzed the impact of LD on QoL by means of more-specific instruments. The aim of this study was to assess the factors associated with a greater impact on QoL in a sample of patients with LD.
PATIENTS AND METHODS
Subjects. Patients who had reported body changes in a previous study were selected for this analysis. Patients included were receiving stable antiretroviral treatment (ART) and had no opportunistic infections or significant decrease in performance status, had a history of >1 year of HAART, and were able to understand the questionnaire. Patients with clinical LD were asked to provide their consent to participate in this new study. The Ethics Committee of Hospital Clínic Universitari de Barcelona had approved the previous study.
The patient characteristics used for the analysis have been reported elsewhere. In brief, they included age, sexual orientation, marriage or partnership status, socioeconomic status, employment status, personal history of psychiatric diagnoses or psychiatric treatment, self-rated physical and psychological status, data on HIV 1 infection and ART, and data on lipodystrophic changes in several parts of the body.
Measurement of QoL. To measure QoL, patients completed the Spanish adaptation of a modified form of the Dermatology Life Quality Index (DLQI). Because there are no specific methods to measure QoL for patients with LD, we decided to use the DLQI, because it was designed to measure the effect of skin diseases on QoL. Many skin diseases, such as atopic eczema, psoriasis, and acne, may produce a disabling external image and a stigmatizing effect and, like LD, may be detrimental to QoL. The DLQI consists of questions on the influence of skin problems on several factors during the 7-day period before completion of the questionnaire: feeling ashamed, problems in daily activities, influence on dressing, impact on spare-time activities, impact on sports participation, impact on work or school, problems in interpersonal relationships, sexual problems, and attempts to solve the problems as a result of body changes. Each question includes a Likert-type scale with 4 answers: 0 (no problems), 1 (few problems), 2 (many problems), and 3 (very many problems). Although the original questionnaire asks about the impact of skin changes on several dimensions of QoL, for the purpose of our study, we made a minor modification, replacing the term "skin problems" with "body changes."
Statistical analysis. First, univariate and multivariate linear regression analyses were performed to predict which patient characteristics and which lipodystrophic changes were associated with an impairment of QoL due to body changes, as measured by the DLQI. Each item of the DLQI was considered to be a dichotomous dependent variable, which comprised the presence of problems as a result of body changes (i.e., a score other than 0 in that item) and the absence of problems (a score equal to 0).
Later, for patients with LD, proportional odds model analysis was performed to explore potential covariates associated with the degree of QoL impairment due to body changes. The dependent variable was the total DLQI score divided into 4 categories: mild impairment (score range, 02.5), moderate impairment (score range, 2.55), substantial impairment (score range, 59), and severe impairment (score, >9). Variables with a significant P value in the univariate model were entered in the multivariate model to detect which patient characteristics or factors could predict a poorer DLQI score.
For all regression analyses, variables with a P value <.1 in the univariate analysis were selected and entered in the multivariate analysis using a stepwise procedure. All P values were 2-tailed, and statistical significance was set at .05. All calculations were performed with Stata, version 6.0 (StataCorp).
RESULTS
Of 150 patients (105 men and 45 women) enrolled into the earlier study, 84 (56%) reported lipodystrophic changes. Twelve patients (14%) reported changes in only 1 part of the body, and 72 patients (86%) reported changes in 2 parts of the body. Lipodystrophic changes included thinning of the face in 49 patients (58%), increased dorsocervical fat in 5 (6%), breast enlargement in 17 (20%), thinning of the arms in 42 (50%), abdominal enlargement in 50 (60%), thinning of the legs in 66 (79%), and reduction of the buttocks in 61 (73%). When patients were asked to specify the cause of LD, 39 (46%) attributed it to ART alone; 19 (23%) attributed it to ART plus other host factors, such as diet, stress, or age; 15 (18%) attributed it to ART plus HIV-1 infection; 5 (6%) attributed it to HIV-1 infection exclusively; and 6 (7%) attributed it to host factors only.
Table 2 lists the proportion of patients who reported problems associated with each item of the DLQI and the significant association of each problem with patient characteristics, HIV-1 infection, ART, and LD-associated changes in buttocks, abdomen, arms, breast, neck, face, and legs. Patients (20-65%) reported problems with: feeling ashamed of body changes; problems with daily activities; influences on dressing; difficulties in social activities or spare time; difficulties in practicing sports; difficulties at work or school; problems with partner, friends, or family; problems with sex life; and making attempts to solve problems due to body changes. Women reported more problems associated with performance of daily activities (P = .036), dressing (P = .028), and sexuality (P = .028) than did men. Injection drug users reported more difficulties associated with sports participation (P = .002) and work (P = .033) than did patients infected with HIV-1 by other routes. Experience of more non-LD-based side effects was associated with impairment of nearly all dimensions of QoL. Patients with abdominal obesity felt more ashamed of their body changes (P = .043) and had more problems with dressing (P = .049) than did other patients. Patients with changes in the breast reported more problems associated with dressing (P = .041), social or spare-time activities (P = .002), and sexuality (P = .011) than did patients without such changes. Patients with changes in the neck reported more problems associated with sexuality (P = .027) than did patients without changes in the neck. Patients with changes in the face made more attempts to solve problems due to body changes (P = .012) than did patients without these changes. Results of multivariate logistic analysis of each item on the DLQI are shown: females reported problems in daily activities and body changes have influenced dressing; problems in sexual life were reported; problems related to employment were reported; people with psychiatric history reported problems with daily activities; IDUs reported difficulties participating in sports activities;
To analyze the overall impairment of psychosocial functioning due to body changes, a proportional odds model analysis was performed for all patients with LD, with the score of the whole DLQI scale used as dependent variable. Univariate factors, such as time since HIV-1 diagnosis (P = .014), infection due to injection drug use (P = .048), report of more ART side effects (P < .001), and experience of lipodystrophic changes in the breast (P = .003), the face (P = .041), or the legs (P = .039), were significantly associated with greater impairment of psychosocial functioning. In the multivariate analysis, only severity of side effects not associated with LD (P = .008) and experience of lipodystrophic changes in the breast (P = .001) remained statistically significant.
DISCUSSION
Few studies of the impact of LD on QoL have used standardized questionnaires. One study collected data through a survey of participants in an Internet news group involving a discussion of experiences with LD. The most frequently reported psychological and social effects of LD were erosion of body image and self-esteem, problems in social and sexual relationships, loss of locus of control, anxiety related to disclosure issues, and propensity toward demoralization and depression. However, our study has a number of limitations. First, respondents were self-selected and so may not have been representative of the whole population. Second, standardized questionnaires were not used to measure QoL. Third, data on the status of HIV-1 infection in the patients were missing.
Another study, in which visual analog scales instead of standardized questionnaires were used, assessed the impact of LD-associated body changes on different domains of QoL in a sample in which almost all patients (86%) were men. All patients reported that 1 domain was affected by LD. Body changes had an impact on social relationships for 63% of patients, 68% reported an impact on performance of daily activities, 68% reported an impact on sexuality, and 83% reported an impact on self-esteem. Impaired sexuality was correlated with changes in the abdomen and neck; decreased self-esteem was correlated with changes in the abdomen; impairment of social relationships was correlated with changes in the neck, legs, and breast; and impaired performance of daily activities was correlated with changes in the legs and arms.
Other studies have focused on the impact of LD on particular dimensions of QoL, such as sexuality. One study sought to determine whether sexual dysfunction was associated with particular components of HAART and to identify risk factors for the development of sexual dysfunction in patients receiving HAART. They observed that individuals receiving a protease inhibitorcontaining therapy who reported signs of LD were more likely to report sexual dysfunction than were those without the disorder, but they were unable to conclude whether the sexual dysfunction was the result of feeling less attractive because of LD or because of the treatment with protease inhibitors.
In a cross-sectional study of 117 HIV-1infected men who have sex with men and are receiving HAART, those who reported LD (50% of the subjects) were asked to compare sexual behavior and feelings of well-being during the period of the syndrome with such experiences during the preceding period. Decreased sexual enjoyment and activity were observed, although without changes in the frequency of condom use. Subjects also felt less physically well and were less confident in their relationships during their experience of LD. However, a decrease in sexual activity was also reported by patients without LD. Therefore, Dukers et al. could not determine whether the decreases were caused by LD, other aspects of HAART, or both.
Although in our previous study we found no differences in overall QoL between patients with and patients without LD, we noted that some patient characteristics were associated with a greater impact on QoL. Among men who have sex with men, unemployed patients, and patients currently undergoing psychiatric treatment, the decrease in some of the QoL subscale scores associated with psychological well-being was greater for those with LD than for those without LD. We concluded that these groups of patients seem to be more vulnerable to the psychological impact of LD on QoL.
From a study of other chronic disabling illnesses, we know that it is not just the illness itself that determines the nature of emotional response; what the illness means to patients also colors their experience of it. In a study on the presence and severity of psychological morbidity in 115 patients with face and neck burns, women and younger patients showed more psychological problems in adjusting to their injuries. In a sample of 56 patients with burn injury and 26 patients with digital amputation, female patients with cosmetic disfigurement were likely to reduce their activities and social relationships and were prone to pessimism, regardless of the severity of burn injury or the degree of physical functioning after digital replantation.
The impact of body changes on some aspects of QoL involving performance of daily activities such as dressing, sexuality, and practicing sports was greater in women and patients who acquired HIV-1 infection through injection drug use. Multivariate analysis showed that the risk of problems associated with performance of daily activities was 4 times higher among women than among men.
The presence of other side effects in addition to changes in specific parts of the body was also significantly associated with poorer psychosocial function. Certain changes seem to be responsible for an impairment in specific dimensions of psychosocial functioning. For instance, patients with abdominal obesity felt more ashamed of body changes and had more problems with dressing, whereas patients with changes in the breast area had more problems associated with dressing, social and spare-time activities, and sexuality. Patients with changes in the neck reported more problems associated with sexuality than did other patients, whereas those with changes in the face were more likely to try to solve the problems caused by such changes. Results of multivariate analysis indicate that the risk of problems associated with social and spare-time activities is nearly 6 times greater for patients with lipodystrophic changes in the breast area than for patients without these changes and that patients with lipodystrophic changes in the face are 3 times more likely to try to find solutions to their body changes than are patients without facial changes. This finding may be explained by the fact that changes in the face are noticed more by others than are other body changes and, therefore, that patients with facial changes are more worried about stigmatization. Unfortunately, because of the limitations of the instrument we used, we do not know what steps the patients took to alleviate or solve the problems due to body changes.
Finally, patients who were infected with HIV-1 through injection drug use, those who had had HIV-1 infection for a longer duration, those with more-severe nonLD-associated side effects, and those with changes in the legs, the breast, or the face reported greater overall impairment of functioning. However, only self-rated severity of side effects and experience of lipodystrophic changes in the breast remained statistically significant after the multivariate analysis.
In summary, these results corroborate those of our previous study: certain subgroups of patients show greater psychosocial impairment of specific areas of QoL due to body changes. Medical, psychological, and social interventions targeted at these subgroups should be developed to enhance QoL in HIV-1infected patients receiving ART.
|
|
| |
| |
|
|
|
