Letter to the editor: Insufficient Reporting of Randomization Procedures and Unexplained Unequal Allocation: A Commentary on "Dairy-Based and Energy-Enriched Berry-Based Snacks Improve or Maintain Nutritional and Functional Status in Older People in Home Care

Dietary advice with or without oral nutritional supplements for disease-related malnutrition in adults

BACKGROUND

Disease-related malnutrition has been reported in 10% to 55% of people in hospital and the community and is associated with significant health and social-care costs. Dietary advice (DA) encouraging consumption of energy- and nutrient-rich foods rather than oral nutritional supplements (ONS) may be an initial treatment.

OBJECTIVES

To examine evidence that DA with/without ONS in adults with disease-related malnutrition improves survival, weight, anthropometry and quality of life (QoL).

SEARCH METHODS

We identified relevant publications from comprehensive electronic database searches and handsearching. Last search: 01 March 2021.

SELECTION CRITERIA

Randomised controlled trials (RCTs) of DA with/without ONS in adults with disease-related malnutrition in any healthcare setting compared with no advice, ONS or DA alone.

DATA COLLECTION AND ANALYSIS

Two authors independently assessed study eligibility, risk of bias, extracted data and graded evidence.

MAIN RESULTS

We included 94, mostly parallel, RCTs (102 comparisons; 10,284 adults) across many conditions possibly explaining the high heterogeneity.  Participants were mostly older people in hospital, residential care and the community, with limited reporting on their sex. Studies lasted from one month to 6.5 years.  DA versus no advice - 24 RCTs (3523 participants) Most outcomes had low-certainty evidence. There may be little or no effect on mortality after three months, RR 0.87 (95% confidence interval (CI) 0.26 to 2.96), or at later time points. We had no three-month data, but advice may make little or no difference to hospitalisations, or days in hospital after four to six months and up to 12 months. A similar effect was seen for complications at up to three months, MD 0.00 (95% CI -0.32 to 0.32) and between four and six months. Advice may improve weight after three months, MD 0.97 kg (95% CI 0.06 to 1.87) continuing at four to six months and up to 12 months; and may result in a greater gain in fat-free mass (FFM) after 12 months, but not earlier. It may also improve global QoL at up to three months, MD 3.30 (95% CI 1.47 to 5.13), but not later. DA versus ONS - 12 RCTs (852 participants) All outcomes had low-certainty evidence. There may be little or no effect on mortality after three months, RR 0.66 (95% CI 0.34 to 1.26), or at later time points. Either intervention may make little or no difference to hospitalisations at three months, RR 0.36 (95% CI 0.04 to 3.24), but ONS may reduce hospitalisations up to six months. There was little or no difference between groups in weight change at three months, MD -0.14 kg (95% CI -2.01 to 1.74), or between four to six months. Advice (one study) may lead to better global QoL scores but only after 12 months. No study reported days in hospital, complications or FFM. DA versus DA plus ONS - 22 RCTs (1286 participants) Most outcomes had low-certainty evidence. There may be little or no effect on mortality after three months, RR 0.92 (95% CI 0.47 to 1.80) or at later time points. At three months advice may lead to fewer hospitalisations, RR 1.70 (95% CI 1.04 to 2.77), but not at up to six months. There may be little or no effect on length of hospital stay at up to three months, MD -1.07 (95% CI -4.10 to 1.97). At three months DA plus ONS may lead to fewer complications, RR 0.75 (95% CI o.56 to 0.99); greater weight gain, MD 1.15 kg (95% CI 0.42 to 1.87); and better global QoL scores, MD 0.33 (95% CI 0.09 to 0.57), but this was not seen at other time points. There was no effect on FFM at three months. DA plus ONS if required versus no advice or ONS - 31 RCTs (3308 participants) Evidence was moderate- to low-certainty. There may be little or no effect on mortality at three months, RR 0.82 (95% CI 0.58 to 1.16) or at later time points. Similarly, little or no effect on hospitalisations at three months, RR 0.83 (95% CI 0.59 to 1.15), at four to six months and up to 12 months; on days in hospital at three months, MD -0.12 (95% CI -2.48 to 2.25) or for complications at any time point. At three months, advice plus ONS probably improve weight, MD 1.25 kg (95% CI 0.73 to 1.76) and may improve FFM, 0.82 (95% CI 0.35 to 1.29), but these effects were not seen later. There may be little or no effect of either intervention on global QoL scores at three months, but advice plus ONS may improve scores at up to 12 months. DA plus ONS versus no advice or ONS - 13 RCTs (1315 participants) Evidence was low- to very low-certainty. There may be little or no effect on mortality after three months, RR 0.91 (95% CI 0.55 to 1.52) or at later time points. No study reported hospitalisations and there may be little or no effect on days in hospital after three months, MD -1.81 (95% CI -3.65 to 0.04) or six months. Advice plus ONS may lead to fewer complications up to three months, MD 0.42 (95% CI 0.20 to 0.89) (one study). Interventions may make little or no difference to weight at three months, MD 1.08 kg (95% CI -0.17 to 2.33); however, advice plus ONS may improve weight at four to six months and up to 12 months. Interventions may make little or no difference in FFM or global QoL scores at any time point.

AUTHORS' CONCLUSIONS

We found no evidence of an effect of any intervention on mortality. There may be weight gain with DA and with DA plus ONS in the short term, but the benefits of DA when compared with ONS are uncertain. The size and direction of effect and the length of intervention and follow-up required for benefits to emerge were inconsistent for all other outcomes.  There were too few data for many outcomes to allow meaningful conclusions. Studies focusing on both patient-centred and healthcare outcomes are needed to address the questions in this review.

Errors in the implementation, analysis, and reporting of randomization within obesity and nutrition research: a guide to their avoidance

Randomization is an important tool used to establish causal inferences in studies designed to further our understanding of questions related to obesity and nutrition. To take advantage of the inferences afforded by randomization, scientific standards must be upheld during the planning, execution, analysis, and reporting of such studies. We discuss ten errors in randomized experiments from real-world examples from the literature and outline best practices for their avoidance. These ten errors include: representing nonrandom allocation as random, failing to adequately conceal allocation, not accounting for changing allocation ratios, replacing subjects in nonrandom ways, failing to account for non-independence, drawing inferences by comparing statistical significance from within-group comparisons instead of between-groups, pooling data and breaking the randomized design, failing to account for missing data, failing to report sufficient information to understand study methods, and failing to frame the causal question as testing the randomized assignment per se. We hope that these examples will aid researchers, reviewers, journal editors, and other readers to endeavor to a high standard of scientific rigor in randomized experiments within obesity and nutrition research.