Hypertension is a major risk factor for heart disease and stroke. The National Institute for Health and Care Excellence (NICE) and international hypertension guidelines recommend measuring blood pressure (BP) in both arms during initial assessment of high blood pressure. This is not routinely done, and impracticality of the methods may be a reason for this. An inter-arm difference in blood pressure (lAD) is common and is important because: i. If undetected an inappropriately low BP is recorded with consequent underdiagnosis andlor under-treatment of hypertension. ii. lAD is associated with increased risk of cardiovascular morbidity and mortality. iii. Detecting lAD could identify people with peripheral arterial disease (PAD). It is assumed that an lAD is caused by localised upper limb PAD, but no direct evidence exists to support this; existing data is derived from selected groups with established vascular disease. Therefore this thesis by submission of papers aims to I. examine the strength of current evidence for an association of lAD with PAD and cardiovascular disease ii. explore the cross sectional and prospective associations of an lAD with cardiovascular morbidity and mortality in community cohorts representative of primary care populations, and iii. examine the utility of a single sequential pair of BP readings in predicting the presence or absence of a significant lAD measured by reference standard techniques. My systematic review findings showed that a systolic lAD ≥15mmHg was associated with PAD, pre-existing cerebrovascular disease and increased cardiovascular and all-cause mortality. These associations are confirmed in subsequent analyses of new cohorts representative of primary care populations with hypertension, diabetes, and in a large community cohort free of clinically evident vascular disease. Cross sectional associations with markers of vascular stiffness are examined and the aetiology of an lAD is explored. A single pair of measurements is shown to reliably exclude the presence of an lAD confirmed by the reference standard technique, suggesting that a pragmatic measurement technique could initially be employed in primary care. The implications of these findings are discussed and future areas of research identified.