Perceived Impact of Limb Amputation on Sexual Activity: A Study of Adult Amputees

Gail M. Williamson
University of Georgia
Andrew S. Walters
University of California, San Francisco

Although research amply indicates that limb amputation poses serious threats to psychological well-being, little is known about the role of sexual activity in this process. In this study of 76 amputees, three fourths of the participants reported that amputation had restricted their sexual activities to some extent. Among the variables predicting more negative impact on sexual activity were older age, being unmarried, and greater feelings of amputation-fostered selfconsciousness in intimate situations. Higher levels of perceived negative impact of amputation on sexual activity, in turn, were related to more symptoms of depression, and in fact, impact on sexual activity emerged as the most consistent predictor of depressive symptomatology. Few participants (less than l0%) reported having received advice from health care practitioners about how amputation might affect their sexual activity. Results are discussed in terms of implications for interventions aimed at improving adjustment to limb amputation.

Approximately 160,000 amputations are performed in the United States each year, the majority of which involve removal of a lower extremity (e.g., Bradway, Malone, Racy, Leal, & Poole, 1984; Goldberg, 1984). Amputations are done for a variety of reasons, including congenital limb deficiency, vascular insufficiency, cancer, and traumatic injury. Thus, attempting to cope with losing a limb is likely to be accompanied by coping with other illness conditions as well. Although amputees face major physical, social, and emotional adjustments, adaptation varies widely between individuals (Schulz, Williamson, & Bridges, 1991; Williamson, 1996; Williamson, Schulz, Bridges, & Behan, 1994). That is, despite the challenges, a substantial percentage of amputees adapt quite well to their disability. However, surprisingly little is known about the factors associated with successfully meeting the demands that follow losing a limb. Virtually no research exists on amputationrelated changes in sexuality and sexual activities (Goldberg, 1984), a situation consistent with the observation that "sexual health remains a neglected area of care in most medical settings" (Schover & Jensen, 1988,p.v).

A substantial body of sociological and feminist literature documents American culture's worship and pursuit of physical perfection (see Wolf, 1991, for an excellent review). Dominant sexual scripts are based on physical attractiveness, youth, heterosexuality, and the absence of physical defect or deformity (Cantor, 1987; Mazur, 1986; Russo, 1987). Sexual scripts exclude or at least, dismiss sexual interest or activity among disabled persons (Mooney, 1995; Rieve, 1989; Romano, 1977) despite reports of continued interest in and enjoyment of sexual activity among these individuals (Schover, 1989; Schulz et al., 1991). That is, regardless of disability or physical trauma that may limit former sexual behavior, sexuality remains a critical component of identity (Cole & Cole, 1983). For example, a panel of individuals with paraplegia (lower body paralysis) reported that if given the choice, they would choose normal sexual functioning over regaining the ability to walk (Medical World News, 1972, as cited in Strong & DeVault, 1994).

Disability may contribute to loss of self-confidence in sexual proficiency and desirability as a sexual partner reactions often exacerbated by insensitive attitudes among health care providers (Labby, 1984). In addition, because people over the age of 60 account for approximately 85% of all amputations (Schulz et al., 1991), many amputees confront an ageist sexual script (Garrison, 1989; Mulligan & Moss, 1991; Mulligan, Retchin, Chinchilli, & Bettinger, 1988; PowerSmith, 1991). However, research indicates that substantial numbers of men and women report sexual interest and availability (O'Donohue, 1987; Schover, 1989) and sexual activity with a partner into older adulthood (George & Weiler, 1981; Hallstrom & Samuelsson, 1990; Mulligan & Palguta, 1991). In our previous research with amputees, older adults spontaneously and comfortably discussed their interest in sexuality and their perceptions of how amputation influenced their sexual activity (Schulz et al., 1991). In this respect, our observations about the importance of sexuality are consistent with those expressed by amputees in other studies (e.g., Postma et al., 1992).

It is not uncommon to find that successfill adaptation to limb amputation is as much or more a function of psychosocial factors than actual physical limitations (e.g., O'Toole, Goldberg, & Ryan, 1985; Schulz et al., 1991; Williamson, 1995; Williamson et al., 1994). For example, bivariate associations generally are not found between indicators of psychological adjustment and aspects of the amputation itself, such as time since amputation (Frank et al., 1984; Rybarcyzk et al., 1992) and type of amputation (Kashani, Frank, Kashani, Wonderlich, & Reid, 1983). Correlations have been observed, however, between emotional distress and pain (Katz, 1992; Lindesay, 1985; Marshall, Helmes, & Deathe, 1992; Pell, Donnan, Fowkes, & Ruckley, 1993), social isolation (Pell et al., 1993; Thompson & Haran, 1983), and a "poor outlook on life" (Froggatt & Mawby, 1981).

Similarly, researchers in at least one study have shown that factors directly related to the amputation itself (e.g., reason for amputation and difficulty with positioning during sex) were unrelated to decreased sexual activity following amputation (Reinstein, Ashley, & Miller, 1978). However, the availability of a supportive spouse or an equivalent partner has repeatedly been shown to predict higher levels of sexual activity among amputees and other adult populations (e.g., Kellet, 1991; Mulligan & Palguta, 1991; Postma et al., 1992; Reinstein et al., 1978).

Consistent with the results of previous research, in the current study, we adopted the theoretical position that psychosocial factors, in addition to physiological factors, play an important role in sexual activity among amputees. Within this framework, several issues were addressed. Of interest was the extent to which amputation may be perceived as negatively affecting sexual activity. We expected to find evidence suggesting adverse impact, which, in turn, was expected to be related to poorer psychological adjustment in the form of depressive symptomatology. In addition, we investigated a variety of demographic and amputation-related variables as predictors of sexual activity and depressed affect, with particular focus on feelings of selfconsciousness in intimate situations.



A convenience sample of participants was recruited from an outpatient amputee clinic and amputee support organizations. Eligibility criteria included being at least one month post-amputation, absence of cognitive deficits or language problems that might invalidate self-report data, and residence in a community setting. Information about the study was made available through the clinic and support organizations to individuals who met these criteria, and they were asked to participate voluntarily. This procedure was adopted in order to maintain confidentiality of medical records and group rosters. Recruiting participants in this manner made it impossible to determine response rates and the extent to which this sample represents the amputee population as a whole. However, as indicated in the following sections, there was considerable variability in terms of demographic and amputation-related factors, suggesting that a relatively heterogeneous sample was obtained.

Most participants (67%) were interviewed, usually in their homes. The remainder of the sample completed a pencil-and-paper survey instrument that assessed the same variables as the structured interview. Compared to those who completed the survey instrument, participants who were interviewed tended to be older, less educated, and more likely to have undergone amputation because of peripheral vascular disease.


Participants (n = 76; 51 men, 25 women) ranged in age from 29 to 84 years. Most (58%) were 55 years of age or older. More than three fourths (76%) were White. With the exception of one Hispanic, all other participants were Black. Most (65%) were married or living as married. Five individuals (6%) had never been married, 14 others (17%) were separated or divorced, and the remaining 7 participants (12%) were widowed. Although we did not specifically address issues of sexual orientation, to our knowledge, all respondents in this sample were heterosexual.

Slightly more than two thirds (68%) were not employed; 36% of the participants were retired, and 15% categorized themselves as disabled. About half (51%) of this sample of amputees had at least a high school diploma.


Three categories of variables were conceptualized as predictors of adjustment to amputation: standard demographics (e.g., gender, age, education, marital status), factors related to the amputation per se, and self-consciousness in intimate situations attributable to the amputation. Two indicators of adjustment, impact of amputation on sexual activity and symptoms of depression, were assessed.

Amputation-related factors. Participants were asked to indicate type of amputation, cause of amputation, total number of amputations, and time since (first) amputation. The most frequent type of amputation involved a lower extremity (76%; 22% above knee, 54% below knee). Of the individuals with arm amputations, 12 were above elbow and 9 were below elbow. Almost half (42%) reported that their amputation was due to vascular disease, and 56% of these patients were diabetic. Other reasons for amputation were trauma (34%), congenital limb deficiency (12%), and cancer (7%). The majority had undergone one amputation, but 20% had undergone two or more procedures. Time since (first) amputation ranged from 1 month to over 50 years, with a mean of 14.6 years.

A single item assessed the extent to which the amputee's stump or prosthesis caused pain or discomfort during sexual activity (0 = never, 4 = always). Relatively few participants (28o) reported that pain or discomfort was ever a problem in this context.

Participants were also asked if they had received any advice from health care professionals about engaging in sexual activities (0 = no, 1 = yes). The few individuals who reported receiving any advice at all (n = 7) were asked how helpful that advice was to them (1 = not helpful, 3 = very helpful). The advice was perceived as helpful or very helpful by six of the seven individuals in this group.

Amputation-fostered self-consciousness. Three questions assessed aspects of self-consciousness in intimate situations as a result of amputation. Using a scale of 1 (not at all) to 3 (definitely), participants indicated if it bothered them to have their spouse or significant other see their stump or prosthesis, if they felt that seeing their stump or prosthesis bothered their spouse or significant other, and if their amputation had made them feel less comfortable in private or intimate situations. Cronbach's alpha for a measure derived by summing responses to these items was low (.49), indicating that the three selfconsciousness questions did not assess a unitary construct. Consequently, these variables were analyzed separately. If respondents indicated being bothered or feeling uncomfortable, an open-ended format was used to ask about reasons for these feelings.

Impact on sexual activity. A fouritem instrument was devised for this study to measure perceived impact of amputation on sexual activity. Participants rated their current sexual activity level (1 = very active, 4 = not at all active) and how much their amputation had restricted sexual activities (0 = never or seldom did this [before amputation], 4 = greatly restricted). Two additional questions were answered on a scale of 1 (very dissatisfied) to 5 (very satisfied): How satisfied would you say you have been with the amount of sexual activity you've had since your amputation?" and "What about the quality of the sexual relations you have had? Have you been satisfied with that?" The latter two items were reverse scored, and the four items were summed to create a composite index, with higher scores indicating more negative impact (possible range 318). Cronbach's alpha for internal reliability was .74. Mean Impact on Sexual Activity score was 9.6, indicating that the average amputee in this sample reported that amputation had resulted in a moderate amount of negative impact on his or her sexual activities.

Symptoms of depression. To measure depressive symptomatology, we employed the Center for Epidemiological Studies-Depression scale (CES-D, Radloff, 1977). The CES-D consists of 20 items scored on a 4point scale (0 to 3) describing frequency of occurrence during the previous week. Higher scores reflect more depressive symptoms. The CES-D has been used successfully in numerous studies of medically compromised individuals (e.g., Schulz & Decker, 1984; Williams & Schulz, 1988; Williamson & Schulz, 1992a, b, 1995a), including amputees (Schulz et al., 1991; Williamson et al., 1994). In the current sample, Cronbach's alpha for internal reliability was .89.

Although we did not assess actual clinical depression, the CES-D has been shown to predict concurrent and future diagnosis of clinical depression (e.g., Lewinsohn, Hoberman, & Rosenbaum, 1988; Roberts & Vernon, 1983; Rohde, Lewinsohn, Tilson, & Seeley, 1990; Schulberg, McClelland, & Burns, 1987) and to discriminate between community-residing adults and those in in-patient psychiatric units (Himmelfarb & Murrell, 1983). Scores of 16 and above are generally believed to indicate that individuals are "at risk" for clinical depression.

Mean CES-D score for the total sample was 11, indicating levels of depression slightly higher than those typically observed in the general population (Ms ranging from 8.6 to 10; e.g., Radloff, 1977). Seventeen individuals (22.4o) were at risk for developing clinical depression, with scores of 16 or above.


One-way analyses of variance (ANOVAs) were conducted to determine associations between categorical predictor variables and the outcome variables of Impact on Sexual Activity and Symptoms of Depression. Neither Impact on Sexual Activity nor Symptoms of Depression varied according to gender or race (all Fs < 0.61, ns). Marital status was related to both impact on sexual activity, F(1,75) = 8.83,p < .004, and symptoms of depression, F(1,75) = 6.34, p < .01. Compared to their married (or living as married) counterparts, amputees who were not married (or living as married) reported more negative impact on sexual activity (Ms = 8.6 and 11.4, respectively) and more symptoms of depression (Ms = 9.0 and 14.9, respectively). Additional analyses revealed, however, that the relation of marital status to depressive symptoms was largely attributable to the impact of amputation on sexual activity. That is, in an analysis of covariance (ANCOVA), Impact on Sexual Activity emerged as a significant covariate, F(1,73) = 12.08, p < .001, and differences in Symptoms of Depression according to marital status were no longer significant, F(1,73) = 2.50, ns. An ANCOVA for differences in Impact on Sexual Activity scores with Symptoms of Depression as the covariate was also conducted. Symptoms of Depression emerged as a significant covariate, F(1,73) = 12.45, p < .001. However, the effect of marital status on Impact on Sexual Activity remained significant, F(1,73) = 4.84, p < .03.

This pattern of results indicates that although depression may play a role in the association between marital status and impact of amputation on sexual activity, depression did not entirely account for observed differences in the impact on sexual activity between amputees who were married (or living as married) and those who were not.

Consistent with previous research (e.g., Schulz et al., 1991; Williamson et al., 1994), depressed affect was not related to the medical condition that led to amputation or type of amputation, both Fs < 2.33, ns. Impact on Sexual Activity scores were differentiated by medical reason for amputation, F(3,73) = 4.00, p < .01, and type of amputation, F(2,74) = 3.48,p < .04. However, these effects appeared to be largely attributable to age. That is, when age was added to the ANOVAs as a covariate, not only was it a significant covariate in both cases (both Fs > 8.94, p < .004), but also, the effects of reason for amputation and type of amputation on Impact on Sexual Activity became nonsignificant (both Fs < 2.46, ns).

Correlations between continuous predictor variables and Impact on Sexual Activity and Symptoms of Depression are shown in Table 1. As predicted, more negative impact on sexual activity was related to higher levels of depressed affect. This finding underscores the importance of identifying factors that may contribute to the negative impact of amputation on sexual activity. More pain during sex and more discomfort in intimate situations also were associated with more depressive symptomatology. However, the association between these variables and depressed affect appeared to be primarily due to their effects on amputation-related impact on sexual activity. Specifically, when Symptoms of Depression were regressed hierarchically onto Impact on Sexual Activity (Step 1) followed by pain during sex and discomfort in intimate situations (Step 2), the latter two variables were no longer significant predictors of depressed affect (both Betas < .20, ns). In this analysis, after controlling for pain during sex and feeling uncomfortable in intimate situations, the effect of Impact on Sexual Activity on Symptoms of Depression remained significant, Beta = .25, p < .03.

To see table 1.

As can be seen in Table 1, Impact on Sexual Activity was related to several variables assessed in this study. Among these were two demographic factors (in addition to marital status, as reported previously). Specifically, less education predicted more negative impact of amputation on sexual activity. The same was true of older age.

Two amputation-related variables also were associated with Impact on Sexual Activity. Less time since (first) amputation predicted more negative impact. In addition, the more frequently one's stump or prosthesis caused pain or discomfort during sexual activity, the more sexual activities were perceived as negatively affected by amputation. Amputation-related pain during sexual activity was only modestly correlated with time since amputation, r = -.21, p < .03.

Two amputation-fostered self-consciousness variables also were related to Impact on Sexual Activity scores. Feeling that one's spouse or significant other was bothered by seeing the stump or prosthesis was associated with more negative impact on sexual activity, as was feeling less comfortable in private or intimate situations.

To summarize, as hypothesized, the extent to which amputation was perceived as negatively affecting sexual activity predicted higher levels of depressive symptomatology. Although several factors (marital status, pain during sex because of amputation, and feeling uncomfortable in intimate situations because of amputation) were related to symptoms of depression in bivariate analyses, multivariate analyses revealed that these associations were mediated by the impact of amputation on sexual activity. Thus, of the variables assessed in this study, effects of amputation on sexual activity emerged as the most consistent predictor of depressive symptomatology. In addition to factors directly related to the amputation (less time since amputation and experiencing amputation-related pain during sex), being unmarried, older, and less educated predicted greater impact of amputation on sexual activity. Likewise, perceived impact on sexual activity was related to feeling that seeing one's stump or prosthesis bothered the spouse or significant other and, more generally, feeling uncomfortable in intimate situations because of one's amputation.

To go to Part II of this study

To see Reference list

To return to Amputee Home Page