Evidence

Raymont et al investigated the prevalence of mental incapacity in medical in-patients in a cross-sectional study of 159 patients in London. They tried to identify risk factors associated with mental incapacity. They estimated that at least 40% of medical in-patients did not have capacity to make decisions about their treatment, and found a strong association between the lack of capacity and increasing age, as well as diminishing cognitive function. More worryingly, they came to the conclusion that incapacity was rarely detected by clinicians or relatives. ^{Reference Raymont, Bingley, Buchanan, David, Hayward and Wessely1} However, the authors noted that there was little conflict between patients and treating physicians with regard to treatment decisions, in contrast to psychiatric settings where such conflict is more common. The researchers used the MacArthur competence assessment tool for treatment (MacCAT-T), which has shown good interrater reliability. Kahn et al published a prospective observational study in an academic medical centre in California. In a sample of 100 patients they found that 70% of patients lacked some type of decisional capacity. A Mini-Mental State Examination score of below 21 was found to be 100% specific in identifying patients without capacity. ^{Reference Kahn, Bourgeois, Klein and Iosif2} The high proportion of patients lacking capacity in this study may reflect the fact that it was conducted in a university hospital, with a high proportion of severely ill patients.

In psychiatry, Owen et al found that 60% of a cohort of psychiatric in-patients showed a lack of capacity to make treatment decisions. The sample size was large, 350 consecutive UK in-patients. Mania and schizophrenia were highly associated with mental incapacity, while only a minority of patients with depression and personality disorder lacked capacity. The study suggested that 39% of informally admitted patients on psychiatric wards lacked capacity, while 14% of detained patients had capacity. ^{Reference Owen, Richardson, David, Szmukler, Hayward and Hotopf3} In a study published in this Journal, Cairns et al confirmed that a lack of decisional capacity with regard to treatment is common, but by no means inevitable, in psychiatric in-patients. In another study of 112 UK patients, mania and psychosis were associated with mental incapacity, as were poor insight, delusions and Black and minority ethnicity if born outside the UK. ^{Reference Cairns, Maddock, Buchanan, David, Hayward and Richardson4}

In a systematic review, Okai et al found 37 papers on mental capacity. Most of these evaluated tools to test capacity and showed high interrater reliability. Psychosis, severity of symptoms, involuntary admission, and treatment refusal were the factors most strongly associated with incapacity in psychiatric in-patient populations. ^{Reference Okai, Owen, McGuire, Singh, Churchill and Hotopf5} Hotopf emphasises that mental capacity assessments have, up to a point, good construct validity. He does, however, identify problems in defining capacity in particular illnesses, such as anorexia nervosa and personality disorder. ^{Reference Hotopf6} In a recent study, my colleagues and I investigated 688 in-patients in North Wales. We asked senior nurses on all wards in an acute hospital trust to judge the capacity of their patients to make basic, as well as complex, decisions. We included medical, surgical, gynaecological, psychiatric and community hospital patients. Overall, 8% of patients lacked capacity to make either basic or complex decisions, and a further 5% lacked capacity to make complex decisions. Lack of capacity, as judged by senior nurses, was most pronounced in medicine (21% of patients), followed by psychiatry (20%). It was least judged to be a problem in surgery (6%). The proportion of patients identified as lacking capacity was much lower than the proportion identified in previous studies where more stringent criteria were used to assess capacity. ^{Reference Lepping, Sambhi and Williams-Jones7} This suggests a persistent underestimation of incapacity by senior ward staff and clinicians, confirming Raymont et al’s findings back in 2004. It would appear that the introduction of the Mental Capacity Act 2005 has not thus far changed the situation significantly.

Ethical and legal implications

There is good evidence that medical and psychiatric patients usually give approval of their treatment once they regain capacity, even if their initial wishes were overridden. ^{Reference Raymont, Bingley, Buchanan, David, Hayward and Wessely1,Reference Owen, David, Hayward, Richardson, Szmukler and Hotopf8} In psychiatry, the retrospective approval rates are no different between patients admitted voluntarily or involuntarily. The only factor that is significantly associated with retrospective approval is regaining capacity. Owen et al concluded that their findings ‘moderate concerns about surrogate decision-making by psychiatrists’. ^{Reference Owen, David, Hayward, Richardson, Szmukler and Hotopf8} It seems that despite relatively good construct validity of capacity assessments, clinicians overestimate patients’ capacity to make treatment decisions. It is unclear whether this leads to worse outcomes. However, we may compromise good outcome by allowing patients who lack capacity to make decisions about their care. This would tend to suggest that the Mental Capacity Act provisions should be used more frequently.

Beauchamp & Childress state that ‘society has a legitimate interest in good outcomes’. ^{Reference Beauchamp and Childress9} They consider it ethically desirable to have systems in place that promote good outcome. The other important question is the purpose of focusing on capacity. The concept of capacity primarily aims to protect autonomous decision-making. ^{Reference Lepping10} We therefore have to ask whether autonomous decision-making is being jeopardised if we allow decisions to be made by people while they lack capacity to do so. This has clear ethical implications as doing so may fail to protect autonomous decision-making, thus violating the very principles that we are trying to protect. In order to avoid more paternalistic practice as a consequence it will be important to emphasise the need to work with patients to optimise their capacity as much as possible. There may be legal implications of accepting decisions made without capacity. There is a risk of accusations of neglect if outcome is undesirable.