Cardiovascular Atlas
About
Menu

China Salt Substitute and Stroke Study (SSaSS)

A large trial in rural China that tested whether replacing ordinary salt with a lower-sodium, potassium-enriched salt reduces strokes. It enrolled people who had had a stroke or who were 60 or older with high blood pressure, in 600 villages. Villages were assigned by chance to the salt substitute or to ordinary salt. Over about 4.7 years there were fewer strokes and fewer deaths in the salt-substitute villages. The results apply to this group in rural China and should not be assumed for other people.

Also called SSaSS · Salt Substitute and Stroke Study

  • Study
  • In this atlas

Overview

Design
Cluster-randomized, parallel-group trial of 600 villages assigned 1:1. The abstract calls it open-label. The registry records allocation as randomized and masking as DOUBLE (investigator and outcomes assessor). The two descriptions are not reconciled. Prevention trial; no drug phase.
Phase
Not applicable
Enrolled
  • 20,996 participants (enrolled count) (China Salt Substitute and Stroke Study, retrieved )
  • 20,995 participants (enrolled count) (Effect of Salt Substitution on Cardiovascular Events and Death, retrieved )
  • 600 villages (randomized count) (Effect of Salt Substitution on Cardiovascular Events and Death, retrieved )
Sponsor
Funder
Registry
  • NCT NCT02092090
Started
(actual) (China Salt Substitute and Stroke Study, retrieved )
Primary completion
(actual) (China Salt Substitute and Stroke Study, retrieved )
Completion
(actual) (China Salt Substitute and Stroke Study, retrieved )
Countries
China (China Salt Substitute and Stroke Study, retrieved )
Care setting
Rural villages in China, 600 of them according to the abstract. (Effect of Salt Substitution on Cardiovascular Events and Death, retrieved )
Registry status
  • “COMPLETED” (U.S. National Library of Medicine (ClinicalTrials.gov), retrieved )

Technical summary

Open-label (abstract), cluster-randomized trial NCT02092090 in 600 villages, assigned 1:1. Registry enrollment is 20996 and the abstract reports 20,995 enrolled; the reason is unknown. Intervention villages: ongoing dietary advice plus a salt substitute of 75% sodium chloride and 25% potassium chloride by mass; control villages: dietary advice at baseline only (registry) and regular salt (abstract). Primary outcome stroke; secondary outcomes major adverse cardiovascular events (49.09 against 56.29 events per 1000 person-years, rate ratio 0.87, 95% CI 0.80 to 0.94) and death from any cause; safety outcome clinical hyperkalemia. Mean follow-up 4.74 years. Funded by the National Health and Medical Research Council of Australia (abstract). PubMed also lists a grant from the UK Medical Research Council (MR/S019669/1). The registry names The George Institute as lead sponsor. The abstract says open-label while the registry records masking as DOUBLE; not reconciled. No results are posted in the registry. Full text was not accessible, so per-arm event counts and denominators are not recorded.

Arms

  • Salt substitute with dietary advice

    Intervention Salt substitution (lower-sodium, potassium-containing salt)

    Villages whose participants used a salt substitute of 75% sodium chloride and 25% potassium chloride by mass (abstract). The registry describes ongoing dietary advice plus a supply of a reduced-sodium added-potassium salt substitute.

  • Regular salt

    Intervention None named

    Villages whose participants continued regular salt of 100% sodium chloride (abstract). The registry records dietary advice at baseline only and an arm type of no intervention. No separate intervention entity is recorded for the comparator.

Findings

Findings are grouped by the claim they support. Each group keeps its own population and follow-up, and each finding keeps its own denominator and time window. Not reported is not zero, and findings from different contexts are not combined.

In plain words

In SSaSS, in rural Chinese villages, people who had had a stroke or were 60 or older with high blood pressure had fewer strokes when their village used a lower-sodium, potassium-enriched salt substitute, with ongoing dietary advice, than when it kept using regular salt. Over a mean follow-up of 4.74 years the stroke rate was 29.14 events per 1000 person-years with the substitute and 33.65 with regular salt (rate ratio 0.86, 95% CI 0.77 to 0.96).

Applies to Recipient not specified
  • People who had had a stroke, or who were 60 years of age or older and had high blood pressure, living in rural villages in China. The registry defines high blood pressure as uncontrolled: systolic 140 mmHg or more on treatment, or 160 mmHg or more without treatment. In the trial 72.6% had a history of stroke and 88.4% a history of hypertension, and the mean age was 65.4 years. No source read gives a minimum age, and the past-stroke group has no stated age floor, so the recipient is recorded as unknown rather than adult. Pregnancy is not mentioned in the sources read.
  • Rural villages in China, 600 of them. The registry names five Northern Chinese provinces and Tibet. A household phone was an entry requirement.
  • Rural China: five Northern Chinese provinces and Tibet (registry description)
  • Mixed ages with a mean of 65.4 years. Age 60 or over is required for people without a past stroke; no age floor is stated for people with one.
  • Gestational age Not reported
  • Postnatal age Not reported
Follow-up Mean follow-up 4.74 years (abstract). The registry gives a time frame of 5 years.

Stroke as an outcome Primary endpoint

  1. Salt substitute with dietary advice arm

    Measure
    Stroke rate (events per 1000 person-years)
    Value
    29.14 events per 1000 person-years
    Reported as
    29.14 events per 1000 person-years
    Interval
    Not reported
    Denominator
    Not reported (basis: Not reported in the abstract: neither person-years nor event counts are given.)
    Time window
    Mean follow-up 4.74 years

  2. Regular salt arm

    Measure
    Stroke rate (events per 1000 person-years)
    Value
    33.65 events per 1000 person-years
    Reported as
    33.65 events per 1000 person-years
    Interval
    Not reported
    Denominator
    Not reported (basis: Not reported in the abstract: neither person-years nor event counts are given.)
    Time window
    Mean follow-up 4.74 years

  3. Salt substitute with dietary advice compared with Regular salt

    Measure
    Rate ratio, salt substitute against regular salt
    Value
    0.86 ratio
    Reported as
    rate ratio 0.86 (95% CI 0.77 to 0.96), P = 0.006
    Interval
    0.77 to 0.96 (95% confidence interval; the abstract does not name the model)
    Denominator
    Not applicable (basis: Comparison of the two per-arm rate rows)
    Time window
    Mean follow-up 4.74 years

Caveats

  • Limitations. One open-label, cluster-randomized trial in rural China. The result applies to people who had had a stroke or were 60 or over with high blood pressure and met the other entry rules, such as no potassium-sparing diuretic, potassium supplement or serious kidney impairment in the household. It does not show the same effect in other groups or places. The tested package is ongoing dietary advice plus the substitute, compared with dietary advice at baseline only (registry). The abstract calls the trial open-label, while the registry records masking as DOUBLE (investigator and outcomes assessor); the two are not reconciled. The registry counts 20996 participants and the abstract 20,995; the reason is unknown. Only the abstract and the registry were read: the NEJM full text returned HTTP 403 and no open copy exists, so event counts, person-years, per-arm intervals and any correction are not recorded.

Sources and citations (7)

  1. Supports Effect of Salt Substitution on Cardiovascular Events and Death

    Publication · Massachusetts Medical Society (N Engl J Med 385(12):1067-1077) · Edition Abstract only, from PubMed, fetched 2026-10-11. The published date is the print issue date; PubMed also gives an electronic publication date of 2021-08-29. The version of record at nejm.org returned HTTP 403. Europe PMC reports the article as not open access and not in PMC. · Published 2021-09-16 · doi: 10.1056/NEJMoa2105675

    Locator
    Abstract, RESULTS, sentence 4
    How it applies
    Abstract of the main report, for the same population and follow-up as this context.
    Retrieved
    Source effective date
    Not reported
    Source state
    Current
    Snapshot hash
    81ecee71f40d1529e615632d4873d8288566233e5de0fa8bfbf25db8c5ace858
    Reuse basis
    Bibliographic metadata and short factual extraction with quotations under 25 words. The Crossref license link points to NEJM terms of use. Full text not accessed; nothing committed.

  2. Supports Effect of Salt Substitution on Cardiovascular Events and Death

    Publication · Massachusetts Medical Society (N Engl J Med 385(12):1067-1077) · Edition Abstract only, from PubMed, fetched 2026-10-11. The published date is the print issue date; PubMed also gives an electronic publication date of 2021-08-29. The version of record at nejm.org returned HTTP 403. Europe PMC reports the article as not open access and not in PMC. · Published 2021-09-16 · doi: 10.1056/NEJMoa2105675

    Locator
    Abstract, RESULTS, sentence 3
    How it applies
    Mean follow-up stated in the abstract.
    Retrieved
    Source effective date
    Not reported
    Source state
    Current
    Snapshot hash
    81ecee71f40d1529e615632d4873d8288566233e5de0fa8bfbf25db8c5ace858
    Reuse basis
    Bibliographic metadata and short factual extraction with quotations under 25 words. The Crossref license link points to NEJM terms of use. Full text not accessed; nothing committed.

  3. Supports Effect of Salt Substitution on Cardiovascular Events and Death

    Publication · Massachusetts Medical Society (N Engl J Med 385(12):1067-1077) · Edition Abstract only, from PubMed, fetched 2026-10-11. The published date is the print issue date; PubMed also gives an electronic publication date of 2021-08-29. The version of record at nejm.org returned HTTP 403. Europe PMC reports the article as not open access and not in PMC. · Published 2021-09-16 · doi: 10.1056/NEJMoa2105675

    Locator
    Abstract, CONCLUSIONS, sentence 1
    How it applies
    The authors conclude for the same population.
    Retrieved
    Source effective date
    Not reported
    Source state
    Current
    Snapshot hash
    81ecee71f40d1529e615632d4873d8288566233e5de0fa8bfbf25db8c5ace858
    Reuse basis
    Bibliographic metadata and short factual extraction with quotations under 25 words. The Crossref license link points to NEJM terms of use. Full text not accessed; nothing committed.

  4. Background Effect of Salt Substitution on Cardiovascular Events and Death

    Publication · Massachusetts Medical Society (N Engl J Med 385(12):1067-1077) · Edition Abstract only, from PubMed, fetched 2026-10-11. The published date is the print issue date; PubMed also gives an electronic publication date of 2021-08-29. The version of record at nejm.org returned HTTP 403. Europe PMC reports the article as not open access and not in PMC. · Published 2021-09-16 · doi: 10.1056/NEJMoa2105675

    Locator
    Abstract, METHODS, sentences 1 to 3
    How it applies
    Design, population, villages and what each arm used.
    Retrieved
    Source effective date
    Not reported
    Source state
    Current
    Snapshot hash
    81ecee71f40d1529e615632d4873d8288566233e5de0fa8bfbf25db8c5ace858
    Reuse basis
    Bibliographic metadata and short factual extraction with quotations under 25 words. The Crossref license link points to NEJM terms of use. Full text not accessed; nothing committed.

  5. Background China Salt Substitute and Stroke Study

    Registry · U.S. National Library of Medicine (ClinicalTrials.gov) · Edition Record last posted 2021-04-08, status verified 2021-04, fetched 2026-10-11. The JSON has no results section. · nct: NCT02092090

    Locator
    protocolSection.armsInterventionsModule.armGroups and interventions[0].description
    How it applies
    Registry description of the two arms: ongoing advice plus the substitute, versus advice at baseline only.
    Retrieved
    Source effective date
    2021-04-08
    Source state
    Current
    Snapshot hash
    7fb59c0de28942b6334a4072776326676f525b7454e1d33a3c60ee6c7c4cf6d3
    Reuse basis
    Registry record, hash only. The ClinicalTrials.gov terms page is rendered by JavaScript and was not read, so the redistribution basis is unverified. The raw JSON is not committed, because that basis is unverified. Public text is quote, following Ben's decision of 2026-10-13 (medical-atlas-core issue 2) that registries allow short quotations with attribution. This source's terms are unverified or reserve all rights, so any quotation stays short.

  6. Background China Salt Substitute and Stroke Study

    Registry · U.S. National Library of Medicine (ClinicalTrials.gov) · Edition Record last posted 2021-04-08, status verified 2021-04, fetched 2026-10-11. The JSON has no results section. · nct: NCT02092090

    Locator
    protocolSection.designModule.designInfo.maskingInfo
    How it applies
    The registry records masking as DOUBLE (investigator and outcomes assessor); the abstract says open-label. Not reconciled.
    Retrieved
    Source effective date
    2021-04-08
    Source state
    Current
    Snapshot hash
    7fb59c0de28942b6334a4072776326676f525b7454e1d33a3c60ee6c7c4cf6d3
    Reuse basis
    Registry record, hash only. The ClinicalTrials.gov terms page is rendered by JavaScript and was not read, so the redistribution basis is unverified. The raw JSON is not committed, because that basis is unverified. Public text is quote, following Ben's decision of 2026-10-13 (medical-atlas-core issue 2) that registries allow short quotations with attribution. This source's terms are unverified or reserve all rights, so any quotation stays short.

  7. Background China Salt Substitute and Stroke Study

    Registry · U.S. National Library of Medicine (ClinicalTrials.gov) · Edition Record last posted 2021-04-08, status verified 2021-04, fetched 2026-10-11. The JSON has no results section. · nct: NCT02092090

    Locator
    protocolSection.designModule.enrollmentInfo.count
    How it applies
    Registry enrollment 20996, against 20,995 in the abstract. Not reconciled.
    Retrieved
    Source effective date
    2021-04-08
    Source state
    Current
    Snapshot hash
    7fb59c0de28942b6334a4072776326676f525b7454e1d33a3c60ee6c7c4cf6d3
    Reuse basis
    Registry record, hash only. The ClinicalTrials.gov terms page is rendered by JavaScript and was not read, so the redistribution basis is unverified. The raw JSON is not committed, because that basis is unverified. Public text is quote, following Ben's decision of 2026-10-13 (medical-atlas-core issue 2) that registries allow short quotations with attribution. This source's terms are unverified or reserve all rights, so any quotation stays short.

In plain words

In the same trial and time window, deaths from any cause were less frequent with the salt substitute: 39.28 events per 1000 person-years against 44.61 with regular salt (rate ratio 0.88, 95% CI 0.82 to 0.95). Death from any cause was a secondary outcome.

Applies to Recipient not specified
  • People who had had a stroke, or who were 60 years of age or older and had high blood pressure, living in rural villages in China. The registry defines high blood pressure as uncontrolled: systolic 140 mmHg or more on treatment, or 160 mmHg or more without treatment. In the trial 72.6% had a history of stroke and 88.4% a history of hypertension, and the mean age was 65.4 years. No source read gives a minimum age, and the past-stroke group has no stated age floor, so the recipient is recorded as unknown rather than adult. Pregnancy is not mentioned in the sources read.
  • Rural villages in China, 600 of them. The registry names five Northern Chinese provinces and Tibet. A household phone was an entry requirement.
  • Rural China: five Northern Chinese provinces and Tibet (registry description)
  • Mixed ages with a mean of 65.4 years. Age 60 or over is required for people without a past stroke; no age floor is stated for people with one.
  • Gestational age Not reported
  • Postnatal age Not reported
Follow-up Mean follow-up 4.74 years (abstract). The registry gives a time frame of 5 years.

Death from any cause Secondary endpoint

  1. Salt substitute with dietary advice arm

    Measure
    Death rate from any cause (events per 1000 person-years)
    Value
    39.28 events per 1000 person-years
    Reported as
    39.28 events per 1000 person-years
    Interval
    Not reported
    Denominator
    Not reported (basis: Not reported in the abstract: neither person-years nor event counts are given.)
    Time window
    Mean follow-up 4.74 years

  2. Regular salt arm

    Measure
    Death rate from any cause (events per 1000 person-years)
    Value
    44.61 events per 1000 person-years
    Reported as
    44.61 events per 1000 person-years
    Interval
    Not reported
    Denominator
    Not reported (basis: Not reported in the abstract: neither person-years nor event counts are given.)
    Time window
    Mean follow-up 4.74 years

  3. Salt substitute with dietary advice compared with Regular salt

    Measure
    Rate ratio, salt substitute against regular salt
    Value
    0.88 ratio
    Reported as
    rate ratio 0.88 (95% CI 0.82 to 0.95), P<0.001
    Interval
    0.82 to 0.95 (95% confidence interval; the abstract does not name the model)
    Denominator
    Not applicable (basis: Comparison of the two per-arm rate rows)
    Time window
    Mean follow-up 4.74 years

Caveats

  • Limitations. Secondary outcome. The limits of the stroke claim apply: One open-label, cluster-randomized trial in rural China. The result applies to people who had had a stroke or were 60 or over with high blood pressure and met the other entry rules, such as no potassium-sparing diuretic, potassium supplement or serious kidney impairment in the household. It does not show the same effect in other groups or places. The tested package is ongoing dietary advice plus the substitute, compared with dietary advice at baseline only (registry). The abstract calls the trial open-label, while the registry records masking as DOUBLE (investigator and outcomes assessor); the two are not reconciled. The registry counts 20996 participants and the abstract 20,995; the reason is unknown. Only the abstract and the registry were read: the NEJM full text returned HTTP 403 and no open copy exists, so event counts, person-years, per-arm intervals and any correction are not recorded.

Sources and citations (6)

  1. Supports Effect of Salt Substitution on Cardiovascular Events and Death

    Publication · Massachusetts Medical Society (N Engl J Med 385(12):1067-1077) · Edition Abstract only, from PubMed, fetched 2026-10-11. The published date is the print issue date; PubMed also gives an electronic publication date of 2021-08-29. The version of record at nejm.org returned HTTP 403. Europe PMC reports the article as not open access and not in PMC. · Published 2021-09-16 · doi: 10.1056/NEJMoa2105675

    Locator
    Abstract, RESULTS, sentence 4 (death)
    How it applies
    Abstract of the main report, for the same population and follow-up as this context.
    Retrieved
    Source effective date
    Not reported
    Source state
    Current
    Snapshot hash
    81ecee71f40d1529e615632d4873d8288566233e5de0fa8bfbf25db8c5ace858
    Reuse basis
    Bibliographic metadata and short factual extraction with quotations under 25 words. The Crossref license link points to NEJM terms of use. Full text not accessed; nothing committed.

  2. Supports Effect of Salt Substitution on Cardiovascular Events and Death

    Publication · Massachusetts Medical Society (N Engl J Med 385(12):1067-1077) · Edition Abstract only, from PubMed, fetched 2026-10-11. The published date is the print issue date; PubMed also gives an electronic publication date of 2021-08-29. The version of record at nejm.org returned HTTP 403. Europe PMC reports the article as not open access and not in PMC. · Published 2021-09-16 · doi: 10.1056/NEJMoa2105675

    Locator
    Abstract, RESULTS, sentence 3
    How it applies
    Mean follow-up stated in the abstract.
    Retrieved
    Source effective date
    Not reported
    Source state
    Current
    Snapshot hash
    81ecee71f40d1529e615632d4873d8288566233e5de0fa8bfbf25db8c5ace858
    Reuse basis
    Bibliographic metadata and short factual extraction with quotations under 25 words. The Crossref license link points to NEJM terms of use. Full text not accessed; nothing committed.

  3. Background Effect of Salt Substitution on Cardiovascular Events and Death

    Publication · Massachusetts Medical Society (N Engl J Med 385(12):1067-1077) · Edition Abstract only, from PubMed, fetched 2026-10-11. The published date is the print issue date; PubMed also gives an electronic publication date of 2021-08-29. The version of record at nejm.org returned HTTP 403. Europe PMC reports the article as not open access and not in PMC. · Published 2021-09-16 · doi: 10.1056/NEJMoa2105675

    Locator
    Abstract, METHODS, sentences 1 to 3
    How it applies
    Design, population, villages and what each arm used.
    Retrieved
    Source effective date
    Not reported
    Source state
    Current
    Snapshot hash
    81ecee71f40d1529e615632d4873d8288566233e5de0fa8bfbf25db8c5ace858
    Reuse basis
    Bibliographic metadata and short factual extraction with quotations under 25 words. The Crossref license link points to NEJM terms of use. Full text not accessed; nothing committed.

  4. Background China Salt Substitute and Stroke Study

    Registry · U.S. National Library of Medicine (ClinicalTrials.gov) · Edition Record last posted 2021-04-08, status verified 2021-04, fetched 2026-10-11. The JSON has no results section. · nct: NCT02092090

    Locator
    protocolSection.armsInterventionsModule.armGroups and interventions[0].description
    How it applies
    Registry description of the two arms: ongoing advice plus the substitute, versus advice at baseline only.
    Retrieved
    Source effective date
    2021-04-08
    Source state
    Current
    Snapshot hash
    7fb59c0de28942b6334a4072776326676f525b7454e1d33a3c60ee6c7c4cf6d3
    Reuse basis
    Registry record, hash only. The ClinicalTrials.gov terms page is rendered by JavaScript and was not read, so the redistribution basis is unverified. The raw JSON is not committed, because that basis is unverified. Public text is quote, following Ben's decision of 2026-10-13 (medical-atlas-core issue 2) that registries allow short quotations with attribution. This source's terms are unverified or reserve all rights, so any quotation stays short.

  5. Background China Salt Substitute and Stroke Study

    Registry · U.S. National Library of Medicine (ClinicalTrials.gov) · Edition Record last posted 2021-04-08, status verified 2021-04, fetched 2026-10-11. The JSON has no results section. · nct: NCT02092090

    Locator
    protocolSection.designModule.designInfo.maskingInfo
    How it applies
    The registry records masking as DOUBLE (investigator and outcomes assessor); the abstract says open-label. Not reconciled.
    Retrieved
    Source effective date
    2021-04-08
    Source state
    Current
    Snapshot hash
    7fb59c0de28942b6334a4072776326676f525b7454e1d33a3c60ee6c7c4cf6d3
    Reuse basis
    Registry record, hash only. The ClinicalTrials.gov terms page is rendered by JavaScript and was not read, so the redistribution basis is unverified. The raw JSON is not committed, because that basis is unverified. Public text is quote, following Ben's decision of 2026-10-13 (medical-atlas-core issue 2) that registries allow short quotations with attribution. This source's terms are unverified or reserve all rights, so any quotation stays short.

  6. Background China Salt Substitute and Stroke Study

    Registry · U.S. National Library of Medicine (ClinicalTrials.gov) · Edition Record last posted 2021-04-08, status verified 2021-04, fetched 2026-10-11. The JSON has no results section. · nct: NCT02092090

    Locator
    protocolSection.designModule.enrollmentInfo.count
    How it applies
    Registry enrollment 20996, against 20,995 in the abstract. Not reconciled.
    Retrieved
    Source effective date
    2021-04-08
    Source state
    Current
    Snapshot hash
    7fb59c0de28942b6334a4072776326676f525b7454e1d33a3c60ee6c7c4cf6d3
    Reuse basis
    Registry record, hash only. The ClinicalTrials.gov terms page is rendered by JavaScript and was not read, so the redistribution basis is unverified. The raw JSON is not committed, because that basis is unverified. Public text is quote, following Ben's decision of 2026-10-13 (medical-atlas-core issue 2) that registries allow short quotations with attribution. This source's terms are unverified or reserve all rights, so any quotation stays short.

Record details

Record id
cardio:ssass
Content hash
6a96af6db76e48c52fcd13b190e58f771bb16cea115e0430c8fef4eeedfb647d