The study is a large scale cluster randomised trial to evaluate the effects of sodium reduction based upon the use of salt substitute on the risk of stroke defined as the occurrence of stroke and stroke deaths. The corresponding null hypothesis that will be tested is that sodium reduction will have no effect upon stroke risk. The secondary objectives are to determine effects of sodium reduction on major vascular events and total mortality.

Lowering sodium intake and the use of salt substitute have been proved to lower blood pressure levels. Effects on the risks of vascular outcomes have not been defined in an adequately powered randomised trial. Rural Chinese are known to consume very large quantities of sodium and to suffer from very high rates of hypertension and stroke. This study is a large scale cluster randomised trial done in rural areas of China. The study will define the effects of a salt substitute-based sodium reduction strategy on the primary outcome of stroke. Secondary endpoints will include major cardiovascular events and total mortality. The study will be conducted in 600 rural villages across five Northern Chinese provinces and Tibet. The study will recruit 35 individuals at elevated risk of stroke from each village for a total of 21,000 participants. The participating villages will be randomised into intervention and control group with 1:1 allocation. Recruitment will prioritise individuals with a history of stroke but also include older individuals with high blood pressure. Follow up is scheduled for 5 years. Individuals in intervention villages will receive repeated dietary advice and a supply of low sodium salt substitute while individuals in control villages will receive dietary advice at baseline only. Every 6 months throughout follow-up each individual will be contacted by phone call to inquire about the occurrence of stroke, hospitalisation for any cause and diagnoses of any other serious illnesses. The interview will be structured and done by individuals masked to the randomised assignment of each individual. For all deaths identified and all events that might possibly be non-fatal strokes a home visit will be made. A series of process indicators including urinary sodium, urinary potassium, blood pressure, knowledge about sodium and use of salt substitute will also be measured on a random sample of at least 20 individuals drawn from a random selection of 60 villages at baseline and every year thereafter.

Inclusion Criteria:

• Disease history
• Prior stroke and/or
• Aged 60 years or over and with uncontrolled high blood pressure (systolic blood pressure >=140 millimeters of mercury (mmHg) at visit if on blood pressure lowering medication; systolic blood pressure>=160 millimeters of mercury (mmHg) if not on blood pressure lowering medication)
• Ownership of a phone by the participant or a household member

Exclusion Criteria:

• Participant or family member is using a potassium-sparing diuretic
• Participant or family member is using a potassium supplement
• Participant or family member has serious renal impairment
• Participant or family member has other reason for concern about use of salt substitute
• Participant eats most meals outside the home
• Participant is not expected to live longer than 6 months from the date of assessment as estimated by the village doctor
• Another family member living in the same household has already been included in the study