Orthostatic hypotension (OH) is a drop in blood pressure (BP) on standing up from a sitting or lying position. The drop in BP can lead to unpleasant symptoms such as dizziness, fainting and falls. It is a common problem, affecting one in five older people and one in three people with Parkinson’s disease.

Doctors usually treat OH without tablets and medicines, instead giving patients advice on drinking more water, wearing compression stockings and performing muscle tensing exercises (known as non-drug therapies). When these non-drug therapies don’t help, patients are offered medication. The two most commonly used drugs for OH are fludrocortisone and midodrine.

Because these treatments are widely used in the NHS, but we do not really know if they work or not, the NIHR commissioned this piece of research in 2018. To find out what the best treatment is, this study compared fludrocortisone and midodrine to non-drug therapies (known as usual care). The trial aimed to recruit 366 participants, aged 18 years and over with OH from up to 20 UK NHS hospital sites. Participants were randomly allocated one of three treatment groups – usual care, fludrocortisone or midodrine. People with OH have told us that symptoms are the most important outcome to them, so we measured any change in symptoms in all three groups. Participants received their allocated treatment for 12 months and were followed up at 3, 6 and 12 months where they completed a falls diary and some study questionnaires in addition to their routine care.

The trial was designed with a 10-month internal pilot. The pilot was designed to judge whether the full trial would be feasible. For the trial to progress from pilot to full trial it had to recruit 64 participants and have fewer than 15% of participants dropping out. Whether participants changed treatments and how well the outcomes were completed were also considered at the end of the pilot, but without any strict criteria. At the end of the pilot only 13 participants had been recruited from four sites. Five other sites were open but did not recruit any participants. For this reason it was decided that the trial was not feasible and it closed early. There were too few participants to judge the drop out rates and whether changing treatments would be a problem.

All of the study outcomes were completed well by participants with very little missing data. Although the falls diaries were not completed as well as the other outcomes. There were too few participants to judge whether treatments were clinically or cost effective. There were also too few adverse events to judge whether any of the treatments were more unsafe than others.

Feedback from sites (including those who did not open) was that COVID-19 caused research and clinical staff to be redeployed to other roles, such as working on COVID-19 studies, or COVID-19 wards, so they were not able to work on this study. COVID-19 also resulted in more telephone and video consultations which made it more difficult to diagnose orthostatic hypotension.

There were aspects of the study protocol which impacted on recruitment. We excluded people who were already on one of the study treatments or had recently tried it. This excluded 52% of eligible participants. A further 10% were excluded because they had a contra-indication to the study drug (meaning it would be unsafe in them). Some sites reported that participants were reluctant to volunteer because there was a chance they would be randomised to the control (non-drug treatments) when they wanted to be randomised to a drug treatment.

The question which this study intended to answer remains an important question but future research will need to consider novel research design to answer it.