Intended for healthcare professionals


Participation in screening for cardiovascular risk by people with schizophrenia or similar mental illnesses: cross sectional study in general practice

BMJ 2003; 326 doi: (Published 22 May 2003) Cite this as: BMJ 2003;326:1122
  1. D P J Osborn, MRC research fellow1 (dosborn{at},
  2. M B King, professor1,
  3. I Nazareth, professor2
  1. 1 Department of Psychiatry and Behavioural Sciences, University College London, London NW3 2PF
  2. 2 Department of Primary Care and Population Sciences, University College London
  1. Correspondence to: D P J Osborn
  • Accepted 11 February 2003


People with severe mental illnesses are at increased risk for physical illness, including cardiovascular disease.1 2 Guidance from the National Institute for Clinical Excellence recommends monitoring of cardiovascular risk factors in such patients and research into relevant interventions.3 Possible causes of this morbidity include diet, smoking, and drug treatment. Health professionals may be less inclined to manage physical health,2 and patients may be uninterested or poorly motivated. Little evidence exists about the acceptability of disease prevention in people with serious mental illness. We hypothesised that such people would be less willing to participate in assessment of cardiovascular risk.

Participants, methods, and results

We invited patients from seven inner London general practices to a cardiovascular risk assessment at their practice. One group comprised all patients with a diagnosis of schizophrenia or other chronic psychosis (excluding primary mood disorders) on their computer record.4 We used a computer to randomly choose a comparison group twice the size without these diagnoses. General practitioners wrote offering an appointment (including a blood test) with a researcher and explaining the 10 year cardiovascular risk score (calculated from age, sex, smoking status, diabetic status, blood pressure, and cholesterol concentrations). Participants would receive all results and interpretations. We did not invite people under 30, over 75, or with pre-existing ischaemic heart disease, as risk scores do not apply.

We sent letters to 224 people with psychosis and 424 people without psychosis. After a week we telephoned up to three times, once outside working hours. If we did not make contact we sent a final invitation. We excluded from the analysis people who had moved away, those we could not reach, and those with no contact with the practice for three years, in accordance with previous findings.5 This left 182 potential participants with psychosis and 313 without psychosis. Limited, anonymous data on non-participants allowed examination of participation rates (table). Last systolic blood pressure was higher in participants—mean difference 9.0 mm Hg (95% confidence interval 5.1 to 13.0).

Factors affecting participation in screening among all eligible patients and those with psychosis

View this table:

We used logistic regression to examine the association between psychosis (dependent variable) and participation—crude odds ratio 0.76 (0.53 to 1.10). Adjustment for age, sex, practice, and systolic blood pressure made negligible difference to the association between psychosis and participation—adjusted odds ratio 0.74 (0.49 to 1.08). The psychosis group consulted their general practitioner more often—mean difference 1.8 (0.8 to 2.9) per year. Increased consultation rate also predicted screening uptake in the total sample (table). Adding consultation rate to the model changed the odds ratio for participation in screening in the psychosis group to 0.65 (0.43 to 0.98). No interaction terms (involving psychosis and age, sex, consultation rate, or smoking) significantly enhanced the logistic model.

Data were available on drug treatment in people with psychosis. Participation was not significantly associated with being on depot drugs, atypical antipsychotic agents, or higher doses (examined as percentage of maximum daily dose and chlorpromazine equivalents).


Many people with psychosis accepted the offer of a cardiovascular risk assessment, providing a valuable opportunity for health education and promotion. Interest in risk assessment was greater than we had assumed. Participation rates were similar to those in other community research involving blood tests. Our negative finding regarding drug types and doses (proxies for severity of illness) indicates that the psychosis group were comparable to people attending psychiatric outpatients. Psychosis was associated with lower uptake of screening uptake only when we included general practice consultation rates in the analysis. This may not be important in practice. Absolute differences in uptake of screening were small. Any reluctance to accept health screening in the psychosis group was offset by increased opportunities provided by more frequent attendance. The smaller subgroup with psychosis who rarely consult a general practitioner will obviously be more difficult to screen, warranting assertive efforts regarding their physical health, perhaps in cooperation with other more involved agencies.


We are grateful to all the participants, their general practitioners, and practice staff.

Contributors: All authors developed the hypotheses, designed the methods, and contributed to writing the paper. MBK and IN supervised the study. DPJO carried out the study; collected, entered, and analysed the data; and wrote the initial draft. DPJO is the guarantor of the study.


  • Funding DPJO was funded by an MRC research fellowship in health services research. Additional funding from the North Central Thames Primary Care Research Network.

  • Competing interests None declared.

  • Ethical approval: Local research ethics committees of the Royal Free Hospital and Camden and Islington Community NHS Trust. All participants provided informed consent. Limited anonymous data obtained on non-participants.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
View Abstract