Although the literature suggests that adjustment to amputation is analogous to a bereavement process, 1,2 there have been few attempts to measure adjustment to amputation with objective standardized instruments. The few studies that have done so have largely found little support for a universal grief reaction. 3
Gallagher and Maclachlan 4 reported that of a sample of 104 patients who had undergone amputation (mean age, 45 years; mean time since amputation, 8 years) 50 (48%) considered something good had happened as a result of amputation. This expression of personal meaning can be seen as evidence that amputation need not be inevitably distressing. More than 64% rated their major concerns as practical matters rather than depressed mood.
Linveh et al 5 reported that limited data are available on relationships between psychological variables and psychosocial adaptation to amputation. In a population of 61 patients (mean age, 55 years; mean duration since amputation, 16 years) they found, using specially designed scales, that active problem solving was associated with less depression and internalized anger, but emotion-focused coping (concentrating on feelings rather than plans for action) was associated with more anxiety and depression and less acceptance of disability.
Frank et al 6 found that overall scores on the Beck Depression Inventory 7 showed their population of amputees (n = 66; mean age, 61 years; mean duration since amputation, 3.6 years) was no different from the general population but that younger patients were more vulnerable to depression.
Fisher and Hanspal 3 found that rates of depression and anxiety were generally low in a population of people at least 1 year after amputation relative to the general population. Katz and Melzack 8 were also unable to find clinically significant levels of distress in a group of 68 established amputees.
However, standardized measures of psychological distress are not always appropriate for a population with physical disabilities, because these instruments have been developed mainly for mental health patient groups and tend to overestimate the prevalence of depression by including items that are confounded with the physical disability. 9 For this reason, instruments such as the General Health Questionnaire (GHQ) 10 have been designed to consider physical symptoms and have been standardized for patients with medical problems.
In this study, the use of a suitably standardized measure of emotional distress has been used to investigate the natural history of emotional problems in the first 6 months after amputation and to assess the effect of counseling.
The participants in this study were consecutive patients attending for their first limb-fitting appointment after amputation. They were asked to complete a questionnaire at this appointment and again 4 to 6 months later, when a follow-up questionnaire was sent by mail. Of the 122 patients who completed the first questionnaire, 58 (48%) responded to the second. These were 44 (76%) men and 14 (24%) women with a mean age of 64 years (range, 23–91) and a mean time from amputation of 2.8 months. Their causes of amputation were dysvascularity and diabetes (76%), trauma (14%), and tumor and infection (10%). The levels of amputation were lower limb (90%) [3 patients (5%) were bilateral amputees] and upper limb (10%).
The patients completed the questionnaire described in Price and Fisher. 11 They were asked whether they experienced problems with their daily tasks, relatives, prosthesis, housing, phantom pain, body image, depression, anger, work, personal relationships, and sex, and whether they wished to receive counseling. They were also asked to complete a 12-item version of the General Health Questionnaire. 10 This has been shown to have validity and reliability in identifying psychiatric ‘caseness’ in medical patients (i.e., those patients with emotional distress in addition to medical problems).
Respondents are asked to state whether in the last few weeks they have experienced less, the same, more, or much more of such difficulties as poor concentration, worry and strain, depression, and lack of self-confidence.
Two methods of scoring were used: scaled (0, 0, 1, 1) and Likert (0, 1, 2, 3). The recommended cutoff for significant emotional distress is a score of 3 on the scaled system of scoring. 10 Likert scoring, having more response points, is preferable as a measure of change, and frequency analysis can provide a commensurate score for ‘caseness’ if necessary. Statistical analysis was carried out using SPSS for Windows 12 in which a ‘t’ statistic was calculated to identify reliable differences between the groups of patients who marked ‘yes’ and ‘no’ to the items in the first part of the questionnaire
Of the participants who responded to the 2nd questionnaire, 18 (31%) had participated in counseling. On initial assessment, 16 (89%) of these reported at least one of the problems listed in Table 1 (mode = 4). Of the remaining 40 who did not request counseling, 24 (62%) cited some of the problems, but the modal number was 1.
Analysis of the GHQ scores showed that, among those who were counseled, 10 (56%) reached the criterion for ‘caseness’ (scaled score 3 or above), compared with 16 (40%) of the uncounseled group. Likert scoring yielded a similar cutoff score for ‘caseness’ of 13. The mean GHQ score for the counseled group was 14.5 compared with 12.6 for the uncounseled group. These are not significantly different scores, but they do indicate that patients who elected to have counseling were more likely to report concern over psychological problems.
At follow-up, the scores of both groups increased; the mean scores were now 17.2 and 13.4, respectively (Table 2). At this point (4–6 months after amputation), 13 (72%) and 19 (48%), respectively, met the ‘caseness’ criterion. Thirty-one (89%) of the counseled patients and 25 (63%) of the uncounseled patients still reported at least one of the problems listed in Table 1, although the modal numbers reduced from 4 to 3 in the counseled group and increased from 1 to 2 in the uncounseled group. The comparable changes in Likert GHQ scores were increases in both groups of 2.7 points and 0.8 points, respectively. These are not significant differences.
There were significant differences between total GHQ scores and whether problems were marked ‘yes’ or ‘no’ in the group of 58 patients taken as a whole. In particular, those patients who reported problems with anger, effect on relatives, and depression were significantly more distressed (had higher GHQ scores) than those who did not record these problems initially. Weaker differences were seen between those who did and did not report concerns about daily tasks, body image, phantom pain, and housing.
At follow-up, significant differences were found on GHQ scores between the patients who did and did not report problems with depression, anger, the prosthesis, and effect on relatives. A smaller difference was found between those who did and did not report difficulties with personal relationships, daily tasks, body image, phantom pain, and work.
This study has attempted to investigate the utility of a standardized measure of emotional distress in a population of newly amputated patients. It seeks to compare the changes in distress scores of a group who elected to have counseling at an early postoperative phase with those who did not and to compare these with reported experiences of other problems.
From the results, it is clear that those who agreed to counseling were a more emotionally vulnerable group, having a higher distress score and containing a higher percentage of patients meeting the ‘caseness’ criterion, thus suggesting valid self-selection. Counseling soon after surgery did not affect the increases in distress experienced by all patients.
Four to six months after surgery, emotional distress was more commonly seen (53% of all the sample meeting the ‘caseness’ criterion, compared with 45% initially) but with no significant difference in increase between the counseled and uncounseled groups.
At both measurement points, differences in emotional distress were found between those patients who did and did not record problems with anger and depression. Thus, the criterion-related validity of the GHQ in this population was suggested by the identification of the differences in these emotional issues rather than practical ones.
Clearly, counseling needs to be available when problems are likely to become evident. In this study, a period of 4 to 6 months after surgery was sufficient to notice an increase in distress. Previous work in this setting has indicated a critical ‘window’ of between 6 and 24 months. 11 Further work needs to be undertaken to invite patients to participate in counseling about 6 months after amputation and then to evaluate the effect of this intervention later. This study found that those who were most distressed initially were those who requested counseling in the early stages. They may have been able to benefit more if they had been encouraged to re-engage with the service at follow-up when their problems became more apparent. Comparison of scores on a suitable measure of distress such as the GHQ could then be used as a validated measure of change.
The 12-item General Health Questionnaire has been used to measure emotional distress in patients in the early weeks after amputation. Those who chose to participate in counseling had higher distress scores than those who did not. Although this intervention did not prevent the expected increase in emotional problems that was apparent some months later, it was able to identify differences in distress experienced by patients recording problems in other areas of functioning and to highlight those patients who might benefit from further intervention. The GHQ has shown evidence of validity in this population and can be used to investigate change following appropriate counseling.
We thank the clinic staff who organized collection of the data, and especially Sue Walker for administrative help. Special appreciation is due to Margaret Piper for preparation of the manuscript.
1. Parkes CM. Components of the reaction to loss of a limb, spouse or home. J Psychosom Res 1972; 16: 343–349.
2. Schoenberg B, Carr AC. Loss of external organs: limb amputation, mastectomy and disfiguration. In: Shoenberg B, Carr AC, Peretz D, Kutscher D, eds. Loss and Grief; Psychological Management in Medical Practice. New York: Columbia University Press; 1970: 119–131.
3. Fisher K, Hanspal RS. Phantom pain, anxiety, depression and their relation in consecutive patients with amputated limbs: case reports. BMJ 1998; 316: 903–904.
4. Gallagher P, MacLachlan M. Positive meaning in amputation and thoughts about the amputated limb. Prosthet Orthot Int 2000; 24: 196–204.
5. Livneh H, Antonak RF, Gerhardt J. Psychosocial adaptation to amputation: the role of sociodemographic variables, disability related factors and coping strategies. Int J Rehabil Res 1999; 22: 21–31.
6. Frank RG, Kashani JH, Kashani SR, Wonderlich SA, Umlauf RL, Ashkanazi GS. Psychological response to amputation as a function of age and time since amputation. Br J Psychiatry 1984; 144: 493–497.
7. Beck AT, Steer RA, eds. Beck Depression Inventory Manual. New York: The Psychological Corporation; 1987.
8. Katz J, Melzack R Pain ‘memories’ in phantom limbs. Review and clinical observations. Pain 1990; 43: 319–336.
9. Williams AC, Richardson PH. What does the BDI measure in chronic pain? Pain 1993; 55: 259–266.
10. Goldberg D, ed. Manual of the General Health Questionnaire. London: nferNelson; 1978.
11. Fisher K, Price M. How does counselling help people with amputations? J Prosthet Orthot 2002; 14: 102–106.
12. Norusis MJ. SPSS for Windows. Chicago: SPSS Inc.; 1994.