Describing the landscape of nutrition or diet-related randomised controlled trials: meta-research study of protocols published between 2012 and 2022

Abstract

Background & Aims: The landscape of nutrition research has changed over the past decades. We aimed to map the landscape of nutrition or diet-related interventions research, using data from randomised controlled trial (RCT) protocols published in the last decade. Methods: This meta-research study examined nutrition or diet-related RCT protocols published in journals indexed in PubMed, Embase, Cinahl, Web of Science, PsycINFO, or the Global Health Database between January/2012 and March/2022. Two reviewers independently screened the titles and abstracts to check eligibility and one reviewer extracted bibliometric information, study characteristics, and research transparency practices such as protocol registration, conflicts of interest and funding disclosure. We also screened the "Instructions for Authors" of journals with publications in our sample to check for endorsement of SPIRIT, TIDieR, and CONSORT reporting guidelines, and we checked if the authors mentioned these reporting guidelines in their paper. Results: The search retrieved 62,319 records, of which 1,068 met the eligibility criteria and were included in the review. The number of published RCT protocols increased annually between 2012 and 2022, with a mean of 161 (range: 155-163) publications/ year. The USA (n = 165; 15.5%) and Australia (n = 137; 12.8%) published the largest number of protocols. Protocols were published in 148 journals, mainly medical journals (n = 518; 48.5%). Among these journals, 50 (33.8%) endorsed SPIRIT, 111 (75.3%) endorsed CONSORT, and four (2.7%) endorsed TIDieR. In 343 (32.1%) publications the authors mentioned SPIRIT, in 297 (27.8%) CONSORT was mentioned, while 20 (1.9%) mentioned TIDieR. Most protocols reported the RCT registration number (n = 1,006; 94.2%) and included statements about conflicts of interest (n = 952; 89.1%) and funding (n = 994; 93.2%). More than one third of protocols focused on adults and elderly participants (n = 350; 32.7%) and most protocols included participants with a specific clinical condition (n = 726; 68.0%). A single nutrition or diet-related intervention (n = 724; 67.8%) was described in most protocols, with "supplementation, supplements or fortification" (n = 405; 37.9%) and "nutrition education, counseling or coordination of care" (n = 354; 33.1%) being the most frequent types of interventions studied. The most frequent primary outcomes reported were related to clinical status (n = 308; 28.8%), nutritional status (n = 247; 23.1%), and frequency or severity of disease (n = 238; 22.3%). The majority of protocols described a single-centre study (n = 838; 78.5%), with two-arms (n = 844; 79.1%), parallel (n = 1014; 94.9%) design, with a superiority framework (n = 755; 70.7%). Conclusions: The number of protocols on nutrition or diet-related trials being published is increasing, indicating the importance of this type of publication. The mention of relevant reporting guidelines by both researchers and journals remains far from ideal. Most protocols assessed supplementation or fortification and nutrition education, counselling or coordination of care interventions, among adults and the elderly.