Randomization and Blinding
Once the questionnaire and initial measurements were completed, participants were randomized to either a control arm, HBPM alone or HBPM plus HH4NM intervention in a 1:1:1 allocation. We utilized a block randomization schema with block sizes of six generated by a blinded statistician using the procedure PLAN in the statistical software package SAS (SAS Institute, Inc., Cary, NC). Study staff and study physicians and investigators interacting with prospective participants during recruitment were blinded to which arm a participant would be randomized. Study staff, who conducted the study visits and instructed participants on study protocols were unblinded at the time of randomization. Study physicians, investigators and the statistician remained blinded until completion of all study visits.