Do people with bipolar disorder get less heart-disease treatment?
People with bipolar disorder in Scottish general practice had more physical illnesses overall, but those with heart disease or high blood pressure were less likely to be prescribed statins and blood-pressure drugs than other patients.
Source
Multimorbidity in bipolar disorder and undertreatment of cardiovascular disease: a cross sectional study
Study at a glance
- Design
- Cross-sectional — Cross-sectional analysis of routine electronic primary-care records from 314 Scottish general practices, comparing adults with a bipolar Read code against all other adults using age- and sex-standardised odds ratios.
- N
- No single analytic N: the source dataset held 1,751,841 registered patients; the adult comparison was 2,582 people with a bipolar code versus 1,421,796 without, and prescribing analyses used the subsets with coronary heart disease or hypertension.
- Population
- Adults (18 and over) permanently registered with Scottish general practices in 2007, with bipolar disorder identified from primary-care diagnostic codes.
- Outcome
- Recorded prevalence of 32 chronic physical conditions and of multimorbidity; among those with coronary heart disease or hypertension, smoking, blood pressure and cholesterol targets, and prescribing of statins, antiplatelets and antihypertensives.
Structured fields used in claim comparison tables when every cited study has a complete layer.
What they did
The researchers used anonymised records from 314 Scottish general practices covering about a third of the population and identified 2,582 adults with a bipolar disorder code. They compared them with 1,421,796 other adults on 32 common chronic physical conditions, standardising for age and sex. Among patients with coronary heart disease or hypertension, they also compared smoking, whether blood-pressure and cholesterol targets were met, and what cardiovascular drugs were prescribed.
What they found
People with bipolar disorder were more likely to have several physical conditions (odds ratio 1.44 for three or more) and had higher recorded rates of 15 of the 32 conditions, including viral hepatitis, constipation, Parkinson's disease, chronic kidney disease and thyroid disease. Yet recorded hypertension and atrial fibrillation were lower and heart disease was not raised, despite more smoking and diabetes, suggesting under-recording. Among those with coronary heart disease, 70.0% of bipolar patients were on a statin versus 74.4% of others, 29.4% were on no antihypertensive versus 15.8%, and they were more often current smokers; a similar undertreatment pattern appeared for hypertension, although blood-pressure and cholesterol target attainment did not differ.
The limits
What it doesn't show
Because the data are cross-sectional, the study cannot show whether lower prescribing leads to worse outcomes, and the authors note that undertreated patients with poor control may already have died and dropped out of the sample. Diagnoses come from routine codes rather than structured interviews, the recorded bipolar rate of 0.2% is well below the roughly 1% expected (so the group is skewed to more severe cases), and a small number of people coded with psychotic depression could not be separated out. Higher rates of conditions like hepatitis or thyroid disease may partly reflect more blood testing in this group rather than more disease.
Key terms
- Multimorbidity
- Having two or more chronic health conditions at the same time.
- Age- and sex-standardised odds ratio
- A comparison of the odds of a condition between two groups after adjusting for differences in their age and sex make-up, so an older group does not look sicker just because it is older.
- Under-recording
- A condition being present but not entered in the medical record, which makes it look rarer than it really is and means it is less likely to be treated.
- Read Codes
- The coded clinical terms UK general practices used to record diagnoses and procedures in electronic records.
- Statin
- A cholesterol-lowering drug recommended for essentially everyone with coronary heart disease, whatever their starting cholesterol.
Flashcards
0 of 11 answers reviewed
Research intelligence for this paper
See its role on concept claims, tensions it is part of, placement history, and related discoveries.
Quiz yourself
Among patients with coronary heart disease, what did the Scottish primary-care study find for bipolar patients compared with others?
Common questions
If people with bipolar disorder smoke more and have more diabetes, why was their heart disease not higher?
The authors argue this points to under-diagnosis or under-recording: their risk factors predict more cardiovascular disease, and other cohorts show higher cardiovascular death, so the lower recorded rates likely reflect missed diagnoses rather than genuinely less disease.
Does lower statin prescribing matter if cholesterol targets were met equally?
Yes for coronary heart disease, because guidelines recommend statins for all such patients regardless of cholesterol level, so not prescribing one is a missed evidence-based treatment. For hypertension the meaning is less clear, as the authors acknowledge.
Why might doctors prescribe less for these patients?
Suggested reasons include diagnostic overshadowing by the mental illness, less regular attendance during mood episodes, GP discomfort managing severe mental illness, and caution about drugs that are dangerous in overdose; the study itself could not test these explanations.
More on Health equity