A cross-sectional analysis of 40 practices in 332,357 adult patients in Lambeth was undertaken. Factors affecting a (clinically coded) diagnosis of FH were investigated by multi-level logistic regression adjusted for socio-demographic and lifestyle factors, co-morbidities, and medications.

The age-adjusted FH % prevalence rate (OR, 95%CI) ranged from 0.10 to 1.11, 0.00–1.31. Lower rates of FH coding were associated with age (0.96, 0.96–0.97) and male gender (0.75, 0.65–0.87), p < 0.001. Compared to a White British reference group, a higher likelihood of coded FH was noted in Other Asians (1.33, 1.01–1.76), p = 0.05, with lower rates in Black Africans (0.50, 0.37–0.68), p < 0.001, Indians (0.55, 0.34–0.89) p = 0.02, and in Black Caribbeans (0.60, 0.44–0.81), p = 0.001. The overall prevalence using Simon Broome criteria was 0.1 %, we were unable to provide ethnic specific estimates due to low numbers.

Lower likelihoods of FH coding (OR, 95%CI) were seen in: non-native English speakers (0.66, 0.53–0.81), most deprived income quintile (0.68, 0.52–0.88), smokers (0.68,0.55–0.85), hypertension (0.62, 0.52–0.74), chronic kidney disease (0.64, 0.41–0.99), obesity, (0.80, 0.67–0.95), diabetes (0.31, 0.25–0.39) and CVD (0.47, 0.36–0.63). 20 % of FH coded patients were not prescribed lipid-lowering medications, p < 0.001.

Inequalities in diagnostic coding of FH patients exist. Lower likelihoods of diagnosed FH were seen in Black African, Black Caribbean and Indian ethnic groups, in contrast to higher diagnoses in White and Other Asian ethnic groups. Hypercholesterolaemia requiring statin therapy was associated with FH diagnosis, however the presence of cardiovascular disease (CVD) risk factors lowered the diagnosis rate for FH.