DISH 1985.

Methods	DESIGN: parallel, randomised, open DURATION: 13 months (56 weeks) NUMBER OF STUDY CENTRES: 4 COUNTRY OF PUBLICATION: USA SPONSOR: grant from National Heart, Lung and Blood Institute; drugs were supplied by the following companies: Ayerst Laboratories, New York; Merck Sharp & Dohme, West Point, PA; Ciba‐Geigy Corp, Summit, NJ; Boehringer Ingelheim Ltd., Ridgefield, CT; USV Pharmaceutical Corp, Tuckahoe, NY; G.D. Searle & Co, Chicago
Participants	WHO PARTICIPATED: people who were previously enrolled in the HDFP treated with antihypertensive drugs and who had sufficiently controlled hypertension. The dietary change on the return of hypertension after withdrawal of prolonged antihypertensive therapy (DISH) included 7 treatment arms; the results of 2 of those arms met the inclusion criteria for this review (hypertensive and obese patients with either dietary intervention or not) SETTING: outpatient clinic MAIN INCLUSION CRITERIA: HDFP participants; DBP ≥ 95 mm Hg on first screening, confirmed by second screening with DBP ≥ 90 mm Hg; patients had received antihypertensive medication for at least 5 years; eligible participants had to be "controlled"; hypertensive persons defined by: no SBP > 180 mm Hg during previous year average DBP < 95 mm Hg during previous year average of last 2 DBP ≤ 90 mm Hg and neither > 95 mm Hg MAIN EXCLUSION CRITERIA: history of congestive heart failure; myocardial infarction; stroke or transient ischaemic attacks; creatinine level ≥ 2.5 mg/dl; ß‐blocker therapy for angina; glucocorticoid therapy for indefinite period NUMBER (dietary intervention vs no dietary intervention): 87 vs 89 participants were randomised GENERAL BASELINE CHARACTERISTICS (dietary intervention vs no dietary intervention): MEAN AGE [YEARS]: 56 vs 57 GENDER [% MALE]: 32 vs 36 NATIONALITY: USA ETHNICITY: black: 62% to 70%, no further information WEIGHT [kg]: 86 vs 90  BODY MASS INDEX [kg/m2]: ‐ SYSTOLIC BLOOD PRESSURE [mm Hg]: 127.6 vs 127.6  DIASTOLIC BLOOD PRESSURE [mm Hg]: 80.9 vs 79.6  COMORBID CONDITIONS: obesity ANTIHYPERTENSIVE TREATMENT [%]: 100 vs 100 DURATION OF HYPERTENSION: at least 5 years in both groups SUBGROUP ANALYSES: results from participants with mild and severe hypertension (not clear whether subgroups were predefined or post‐hoc)
Interventions	LENGTH OF FOLLOW‐UP: 13 months (56 weeks ) DIETARY INTERVENTION: according to revised "Metropolitan Life Insurance" standards; intervention consisted of 8 initial weekly group sessions followed by monthly sessions plus individual consultation as needed NO DIETARY INTERVENTION: no recommendations ADDITIONAL TREATMENT: discontinuation of antihypertensive treatment using a step‐down withdrawal program; biweekly consultations for BP measurement for 16 weeks followed by monthly consultations; no change in salt uptake
Outcomes	PRIMARY OUTCOMES: 1. MORTALITY: ‐ 2. CARDIOVASCULAR MORBIDITY: ‐ 3. ADVERSE EVENTS: Definition: withdrawal due to the need to restart antihypertensive medication SECONDARY OUTCOMES: 1. CHANGES IN SYSTOLIC BLOOD PRESSURE [mm Hg]: ‐ 2. CHANGES IN DIASTOLIC BLOOD PRESSURE [mm Hg]: ‐ 3. BODY WEIGHT [kg]: Definition: body weight change from baseline to endpoint visit ADDITIONAL OUTCOMES MEASURED IN THE STUDY: 1. Number of participants without antihypertensive medication 2. Change in sodium excretion
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Comment: No details on generation sequence are provided
Allocation concealment (selection bias)	Unclear risk	Comment: Method of concealment is not described
Blinding (performance bias and detection bias) All outcomes	High risk	Comment: Open design, no blinding described
Incomplete outcome data (attrition bias) All outcomes	High risk	WITHDRAWALS (dietary intervention vs no dietary intervention): 0 vs 0  concerning the endpoint success of withdrawal from antihypertensive medication; 20 vs 12 concerning body weight at week 56; although there were no losses to follow‐up in the relevant subgroups concerning success of discontinuing antihypertensive treatment, for 13% respectively 23% participants weight at week 56 is not reported REASONS/DESCRIPTIONS: No reasons for incomplete data are mentioned
Selective reporting (reporting bias)	Unclear risk	Comment: No primary and secondary outcomes were defined
Other bias	High risk	Comment: 1. Randomisation before consent of participants (possible selection bias) 2. Different body weight at baseline 3. Participants participated in previous trial for 5 years (selected population may not be representative)