We performed a prospective, longitudinal study on mental health in 299 prostate cancer patients, using validated instruments. At 1 month before treatment roughly one in every four patients was classified as ‘high-anxiety'. After 6 months, after initiation of treatment, men reported significantly less anxiety and feelings of depression and a significantly better mental health. Average scale scores remained at the improved levels through follow-up.
Before interpreting the results from a clinical perspective, two methodological issues have to be discussed. First, there is no unequivocal cutoff value for the STAI to define high-anxiety. We used the earlier reported cutoff value of 45 and higher (Millar et al, 1995
; Roth et al, 1998
) which matches almost perfectly with the validated cutoff value of the HADS (Millar et al, 1995
). Second, nonresponse at 5-year follow-up was present. Although, repeated measures analyses can take incomplete cases into account, we excluded the 5-year assessment in evaluating the STAI-State as a screening tool; because the nonresponse at 5-year follow-up was significantly lower in high-anxiety men, a complete case analysis would not be appropriate. Furthermore, it can be argued that a prediction of high-anxiety at 12-months follow-up is clinically more relevant, because high levels of anxiety or symptoms of depression at 5-year follow-up may be unrelated to the preceding prostate cancer diagnosis and treatment.
When we selected an anxiety measure around 1996, validated Dutch versions of both STAI and HADS were available. We needed only one measure and chose STAI, but, in retrospect, we could have chosen HADS as well. We think that in this context there are no really strong scientific arguments to prefer the one to the other. Both are well-validated and commonly used measures for generic anxiety, each with their own strengths and weaknesses. The average STAI-State scores at baseline in the high-anxiety groups, that is, 52 for surgery patients and 54 for radiotherapy, are high in comparison to, for example, the mean STAI-State score in a group of males with anxiety neurosis of 45 (van der Ploeg et al, 1980
). Of high-anxiety men, prostatectomy patients reported less anxiety and feelings of depression and better mental health at follow-up than radiotherapy patients. A first possible explanation is that the level of anxiety influenced the treatment decision; high-anxiety men may have perceived surgery as too frightening and therefore opted for radiotherapy. A second explanation could be that surgery led to more reassurance since – contrary to radiotherapy – the prostate actually is removed. Previous research has suggested that men may choose surgery on the basis of the lay belief that surgical removal is the most effective way to cure cancer (Steginga et al, 2002
). A third explanation could be age, since high-anxiety radiotherapy men were significantly older than high-anxiety prostatectomy men. However, for a number of reasons, age does not
seem to be the explanation. For instance, in spite of the difference in average age low-anxiety men who were treated by radiotherapy reported similar levels of anxiety and symptoms of depression as surgically treated low-anxiety men. Furthermore, several studies reported an association of higher age and lower levels of anxiety. A review study, for instance, reported some evidence that ageing is associated with an intrinsic reduction in susceptibility to anxiety and depression (Jorm, 2000
). Furthermore, older men reported better mental health, although higher ages were associated with worse physical health (Litwin et al, 2002
). And finally, higher levels of anxiety were found in prostate cancer patients younger than 65 years of age (Lintz et al, 2003
). A fourth possible explanation could be that compared to the high-anxiety surgery group, a higher percentage of men were ‘single' – that is, in most cases divorced or widowed – in the high-anxiety radiotherapy group. It has been reported before that marital status contributes to happiness (Joung et al, 1994
). A fifth possible explanation could be the higher rate of recurrence in men treated by radiotherapy vs
men treated by surgery, that is, 21 vs
5%. Within the radiotherapy group, the recurrence rate did not differ significantly between high- and low-anxiety men.
The frequency of side effects in this same group of patients has been reported elsewhere (Korfage et al, 2005
). Four to five years post-treatment, side effects were reported by higher percentages of men treated by surgery than by men treated by radiotherapy, for example, 88% of erectile dysfunction vs
64%, and 31% of urinary leakage vs
13%. Thus side effects appear not to be the reason for higher anxiety levels in men treated by radiotherapy.
Our findings are in line with a cohort study on 111 prostate cancer patients with 12 months follow-up. Steginga et al (2004)
found that psychological and treatment decision-related distress decreased with time, independent of treatment choice. At 12 months follow-up, most men experienced low levels of distress. The authors suggested that, in general, men are resilient to the experience of localized prostate cancer and adjust well psychologically. We agree that the majority of localized prostate cancer patients seem to do fine in the 1–5 years following treatment, but this is not the case for all patients. The challenge for clinicians is to detect those men early who will experience ongoing clinical levels of anxiety and symptoms of depression, and provide those with in-depth support. A (short) anxiety measure could be a useful tool. We applied a 20-item version of the STAI-State. Currently validated 6-item versions are available in English (Marteau et al, 2001
) and other languages such as Dutch (van der Bij et al, 2003
Vedana and co-workers compared the Hospital anxiety and depression scale (HADS) and STAI to identify the most suitable instrument for screening a population at in-hospital intensive rehabilitation on anxiety and depression. The sensitivity of the STAI (52%) was less than that of the HADS (72%), but its specificity (99%) was greater than that of HADS (84%). The authors concluded that both instruments can be recommended for psychological screening of patients in an in-hospital intensive rehabilitation (Vedana et al, 2002
). A cross-sectional study on psychiatric disorders after successful renal transplantation assessed the value of self-report scales, among others STAI and HADS, in predicting anxiety and depression. HADS was found to significantly (P
=0.003) predict anxiety and depression (Arapaslan et al, 2004
In our study, using the STAI-State baseline score as a screening tool resulted in the early detection of 71, respectively, 60% of patients who were to experience high levels of symptoms of anxiety, respectively, feelings of depression at 6 or 12-months follow-up. The sensitivity might be improved by expanding the tool with, for instance, disease characteristics as the Gleason score, or by applying other measures, for instance the HADS (Zigmond and Snaith, 1983
Treating clinicians may not always realize that, in spite of a comparably favourable prognosis, so many patients experience high levels of anxiety and symptoms of depression after a diagnosis of localized prostate cancer. We recommend clinicians to attempt early detection of patients at risk of such high levels and provide them with psychological support. STAI-State can be a useful screening tool but needs further development.