Half of patients on long-term antipsychotic therapy, both typical and atypical, experience weight gain as a side-effect (Reference BaptistaBaptista, 1999). This is a phenomenon shared with many other psychotropic agents. The tendency of atypical antipsychotics, in particular clozapine and olanzapine, to induce weight gain to a larger extent than that of the conventional agents has renewed interest in this area (Reference Kurzthaler and FleischhackerKurzthaler & Fleischhacker, 2001). It must be noted that there is marked variation between individuals in experiencing weight gain as a side-effect.

Mean weight gains with atypical agents range from 0.04 kg with ziprasidone to 4.45 kg with clozapine at 10 weeks (Reference Allison, Mentore and HeoAllison et al, 1999). Weight gain has been described as ‘one of the most distressing side-effects of antipsychotic medication’ (Reference Fakhoury, Wright and WallaceFakhoury et al, 2001). Increased morbidity and mortality in association with being overweight is well established. In addition, people with schizophrenia have higher rates of morbidity and mortality than the general population (Reference Brown, Barraclough and InskipBrown & Barraclough, 2000).

Abnormalities in glucose regulation occur more commonly in individuals with schizophrenia than in the general population. Treatment with antipsychotic medications, both typical and atypical, is associated with impaired glucose metabolism, exacerbation of existing type 1 and type 2 diabetes, and new onset of type 2 diabetes mellitus and diabetic ketoacidosis (Reference Haupt and NewcomerHaupt & Newcomer, 2001).

It has been suggested that screening measures, including recording baseline weight and monitoring blood glucose control, should be introduced prior to initiating and during prescription of antipsychotic therapy (Reference Kurzthaler and FleischhackerKurzthaler & Fleischhacker, 2001). We aimed to survey current practice in relation to monitoring of weight and blood glucose prior to initiation and during prescription of antipsychotic medication in an in-patient setting.

Method

A list of all patients with an ICD–10 diagnosis of schizophrenia (World Health Organization, 1992), schizoaffective disorder, persistent delusional disorder, acute and transient psychotic disorder or induced delusional disorder admitted to acute general adult wards in The Royal Cornhill Hospital from the beginning of January 2001 to the end of March 2001 was obtained from medical records. In accordance with local policy, consent was sought from the individual patient's responsible medical officer prior to accessing the case notes.

Case notes were reviewed, looking for any recordings of weight, during the index admission and in the year thereafter. A record of blood glucose was sought in both the case notes and from the regional laboratory computer database at the point of admission. If the random glucose was noted to be elevated (levels were deemed to be elevated if greater than or equal to 10 mmol/l), a further search was made for subsequent checks during the admission.

Results

Sixty-one patients with the specified diagnoses were admitted to general adult wards during the 3-month study period. Consent to examine the medical record was obtained from the responsible medical officer for 50 of the cases. Six sets of those patients' case notes were unavailable. Seven of the remaining 44 patients had two admissions to general adult wards during the specified period. Each admission was dealt with individually; this resulted in a data-set of 51 admissions. All patients were prescribed antipsychotic medications.

Weight

Twenty-two of the 51 admissions (43%) had their weight recorded by nursing staff. Three of these weights were noted in the admission medical examination. One additional weight was recorded by the admitting doctor and not by nursing staff. Two patients had comments ‘over-weight’ and ‘slightly overweight’ recorded by the admitting doctor, but no corresponding entry of actual weight by nursing or medical staff. For the purpose of analysis, these two qualitative entries are recorded as ‘weight not taken’. In total, 23 of the 51 admissions (45%) had a record of weight kept (Table 1).

Table 1. Weights recorded among 51 admissions

Weight measurement	n (%)
Weight recorded on admission
By nursing staff	22 (43)
By medical staff	3 (6)
By nursing or medical staff	23 (45)
Weight recorded following admission
At any other time during first admission	1 (2)
In following year	21 (41)

While in-patients, 27 individuals were initiated on, or had a change in, antipsychotic medication. Of these 27, 12 (44%) had a record of admission weight.

Case records were hand-searched for the year following the study period to determine whether further entries of weights had been made. Of those with baseline weights (n=23) only one had repeated weights taken during admission. This patient was commencing treatment with clozapine. Eleven of the cases had a subsequent entry of weight when they were readmitted. Weight changes ranged from a loss of 22.3 kg to a gain of 6.4 kg. Ten admissions without baseline weights had a subsequent entry of weight. Eight of these were on readmission, one at an out-patient clinic and another recorded in a letter from a dietician to whom they had been referred because of weight gain.

Of the seven cases with two admissions, only two had a weight taken on the first admission with none on the second admission. The remaining five had no record taken on either occasion.

Discussion

Weight gain as a side-effect of antipsychotic medication has recently emerged as an area of concern. This side-effect has the potential to impact on both the physical and psychological well-being of the patient. Failure to address this issue might lead to life-threatening medical complications and non-adherence. Evidence of being overweight or having raised blood glucose in a patient taking an antipsychotic agent does not necessarily mean that the antipsychotic has caused this. Evidence of a change in these parameters that coincided with the introduction of the antipsychotic, however, would be more suggestive of a causal relationship. Standardised monitoring would help to clarify these links.

Despite prompts being present on both the nursing and medical admission records, less than half of all admissions had a record of their weight being taken. This remained the case even when patients were initiated on a new antipsychotic regime. These prompts comprised a specific heading for weight at the top of the medical admission physical examination form and a similar cue on the nursing admission form.

This survey suggests that there is no reliable process in place for recording weight during admission and subsequent to it. For the majority of patients who had a subsequent record of weight, this was on their readmission to hospital rather than at out-patient follow-up. Weight gain most frequently occurs in the first 4-12 weeks of treatment (Reference WetterlingWetterling, 2001). However, a further increase has been observed during long-term follow-up (Reference WetterlingWetterling, 2001). This would suggest that regular out-patient monitoring would be beneficial.

Hyperglycaemia as an association of antipsychotic treatment, especially with olanzapine and clozapine, has been highlighted. Blood glucose monitoring for all patients commenced on clozapine and olanzapine has been advocated (Reference Mir and TaylorMir & Taylor, 2001). Further support for this position comes from the Committee on Safety of Medicines warning of an association between potentially fatal ketoacidosis and olanzapine (Committee on Safety of Medicines, 2002).

Routine examination of glucose control on all patients treated with antipsychotic medications has been recommended (Reference Haupt and NewcomerHaupt & Newcomer, 2001). This can be achieved by taking random blood glucose levels, fasting blood glucose, HBAlc monitoring or performing a glucose tolerance test. Random blood glucose is the least reliable of these measures (Reference Haupt and NewcomerHaupt & Newcomer, 2001). Annual checks of fasting blood glucose levels in all patients treated with any antipsychotic medication, as well as assessment of fasting levels prior to commencing an antipsychotic medication, have been advocated (Reference Haupt and NewcomerHaupt & Newcomer, 2001).

In this survey, random blood glucose measurements were the only measurement of glucose taken and these were checked haphazardly. In particular, half of the patients initiated on olanzapine and none of those initiated on clozapine had any form of blood glucose measurement taken. When random levels were found to be elevated (n=2), no fasting levels were taken.

The Royal Cornhill Hospital is a large teaching hospital, and this survey reflects the work of 14 different teams. Our sample is small and it would be helpful to repeat this survey elsewhere with large numbers. After this survey was completed, the Committee on Safety of Medicines (2002) released a warning on antipsychotic-associated hyperglycaemia with olanzapine; monitoring of this parameter may have improved as a consequence of this warning.

This survey appears to provide evidence of the need for standardisation of weight measurement and blood glucose monitoring in this vulnerable group of patients. Such monitoring would allow for early management of side-effects and/or a change in medication.

Declaration of interest

None.