HINTS.

Study characteristics
Methods	Study design: parallel, 4‐arm RCT; randomization stratified by diabetic status Unit of randomization and analysis: individual Number randomized: total: 252 diabetics of 593 participants (diabetics and non‐diabetics) Per group: (diabetics only) 49 usual care, 49 medication management, 50 behavioral management, 47 medication and behavioral management Number analyzed: 48 usual care, 45 medication management, 50 behavioral management, 43 medication and behavioral management Sample size calculation: "We estimated the necessary sample size empirically through a simulation study. We expect approximately 15% of the enrolled sample to dropout by the end of the study based on our prior studies. For a type‐I error of 0.05 and 80% power, we require a total of 600 patients to be able to detect a 15% increase in the probability of BP control" for the overall cohort. Length of follow‐up (planned and actual): 18 months Number of arms: parallel, 4‐arm RCT; randomization stratified by diabetic status
Participants	Country: USA Study period: May 2006 to August 2010 Type of diabetes: not reported; however, given age of participants probably type 2 Participants’ status at baseline: Age, year, mean (SD): usual care 64 (11); medication management 63 (11); behavioral management 64 (10); medication and behavioral management combination 63 (11) Gender, % women: usual care 2%; medication management 6%; behavioral management 0%; medication and behavioral management combination 8% Race/ethnicity): % white: usual care 47%; medication management 42%; behavioral management 40%; medication and behavioral management combination 53% Duration of diabetes, years, mean (SD): not reported Smokers, % current: parent study 22% usual care, 19% behavioral only, 20% medication only, 22% medication and behavioral combined Blood pressure, mean (SD): not reported for DM subgroup. Required > 140/80 for eligibility Parent study, "controlled BP" SBP 128 usual care, 128 behavioral only, 130 medication and behavioral combined, mmHg DBP 78 usual care, 77 behavioral only, 78 medication and behavioral combined, mmHg Hypertension, %: 100% Glycated hemoglobin, mean (SD): not reported BMI, mean (SD): parent study 30.4 medication only or medication and behavioral combined, 30.0 usual care, 30.6 behavioral only Severity of diabetic retinopathy: not reported at baseline; within 2 months of study enrollment, 31% usual care, 32% behavioral, 27% medication and behavioral Medical history: Myocardial infarction, %: 0% (within 3 months) Stroke, %: 0% (within 3 months) Equivalence at baseline: groups appear similar at baseline Inclusion criteria: military veterans diagnosed with uncontrolled blood pressure (defined as average > 140/80 mmHg), prescribed medicine to lower BP, regular primary care physician at Duke VAMC. For the DM subgroup, dilated examinations of the eyes had to have been performed at baseline (before or within 2 months after) and at 1 or more follow‐up visits 365 days or more after enrollment. Exclusion criteria: hospitalization for stroke in past 3 months; heart attack (unclear whether ever or past 3 months); surgery for blocked arteries (same comment); diagnosed with metastatic cancer or treated with dialysis, diagnosed with dementia or hearing impairment that prevents hearing/speaking by telephone; creatinine serum lab value > 2.5 mg/dL
Interventions	Intervention 1: usual care (no intervention) Intervention 2: behavioral: nurse behavioral intervention with home BP telemonitoring Intervention 3: medication: nurse medication management with home BP telemonitoring Intervention 4: combined behavioral and medication: nurse administered tailored behavioral and medication management Length of follow‐up: Planned: 18 months Actual: 18 months
Outcomes	Primary outcome, as specified for this review: progression of retinopathy from absence to presence or less severe to more severe from the time of enrollment to the most recent follow‐up 365 days or later (progression at 1 to 2 years after randomization) Secondary outcome, as specified for this review: none reported for DM subgroup Other diabetic retinopathy outcomes: none mentioned Retinopathy diagnosis and monitoring: chart review by single abstractor Participants (eyes) examined for the outcome: 194 total (numbers reported by arm sum to 196) Intervals at which outcomes were assessed: baseline and 6‐month intervals from enrollment; for DR analysis, at least once at 365 days or longer after enrollment Cost of interventions: not reported Quality of life: not reported Adverse events: not reported Other outcomes reported from the study: the primary study reported change in BP from baseline at 6‐month intervals
Notes	Source of funding: US Department of Veterans Affairs Declaration of interest: none reported Run‐in length: none Class(es) of antihypertensive agents: not reported in available information Degrees of blood pressure control achieved: in parent trial, investigators reported  “moderate reduction in systolic BP achieved HINTS via the nurse‐administered telemedicine program (8 mmHg lower in the combined intervention group vs the control group" Trial registration number: NCT00237692 Other: after controlling for duration of follow‐up, the odds of diabetic retinopathy progression were significantly greater among participants receiving usual care than among participants receiving medication management, either alone or in combination with behavioral management (OR 2.16, 95% CI 1.03 to 4.52; P = 0.04), but did not differ from the group receiving behavioral management alone (OR 0.88, 95% CI 0.40 to 1.95; P = 0.84).
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	"Consenting patients are then randomized to 1 of the 4 arms using consecutively numbered envelopes; randomization is stratified by diabetic status"
Allocation concealment (selection bias)	Low risk	“consecutively numbered envelopes, stratified by diabetic status”
Masking (performance bias and detection bias) Primary outcomes	High risk	Per ClinicalTrials.gov: “Masking: None (Open Label)”
Masking (performance bias and detection bias) Secondary outcomes	High risk	Per ClinicalTrials.gov: “Masking: None (Open Label)”
Incomplete outcome data (attrition bias) Primary outcome	Low risk	Diabetic retinopathy was not specified as an outcome of the parent trial.
Incomplete outcome data (attrition bias) Secondary outcomes	Unclear risk	No secondary outcomes reported for the DM group.
Selective reporting (reporting bias)	Low risk	Diabetic retinopathy prevalence at baseline and progression during follow‐up reported only for diabetics in study population.
Other bias	Unclear risk	“a single chart abstractor determined the presence and severity of diabetic retinopathy at baseline and the most recent follow‐up, as recorded in the electronic chart”