Steiber AL, León JB, Hand RK, Murphy WJ, Fouque D, Parrott JS, Kalantar-Zadeh K, Cuppari L. Using a web-based nutrition algorithm in hemodialysis patients.
J Ren Nutr 2014;
25:6-16. [PMID:
25193109 DOI:
10.1053/j.jrn.2014.06.005]
[Citation(s) in RCA: 3] [Impact Index Per Article: 0.3] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/06/2013] [Revised: 03/15/2014] [Accepted: 06/12/2014] [Indexed: 11/11/2022] Open
Abstract
OBJECTIVES
The purpose of this study was to test the ability of a newly developed nutrition algorithm on (1) clinical utility and (2) ability to capture patient outcomes.
RESEARCH DESIGN
This was a prospective observational study, using a practice based research network structure, involving renal dietitians and hemodialysis [HD] patients.
SETTING
This study took place in HD outpatient units in five different countries.
SUBJECTS
Hundred chronic HD patients were included in this study. To select subjects, dietitians screened and consented patients in their facilities until 4 patients "at nutrition risk" based on the algorithm screening tool were identified. Inclusion criteria were patients aged older than 19 years, not on hospice or equivalent, able to read the informed consent and ask questions, and receiving HD.
MAIN OUTCOME MEASURE
The ability of the algorithm screening tool is to identify patients at nutrition risk, to guide clinicians in logical renal-modified nutrition care process chains including follow-up on relevant parameters, and capture change in outcomes over 3 months. Statistics were performed using SPSS version 20.0 and significance was set at P < .05.
RESULTS
One hundred patients on HD, enrolled by 29 dietitians, were included in this analysis. The average number of out-of-range screening parameters per patient was 3.7 (standard deviation 1.5, range 1-7), and the most prevalent risk factors were elevated parathyroid hormone (PTH; 62.8%) and low serum cholesterol (56.5%). At the initial screening step, 8 of the 14 factors led to chains with nonrandom selection patterns (by χ(2) test with P < .05). In the subsequent diagnosis step, patients diagnosed within the insufficient protein group (n = 38), increased protein intake by 0.11 g/kg/day (P = .022). In patients with a diagnosis in the high PTH group, PTH decreased by a mean of 176.85 pg/mL (n = 19, P = .011) and in those with a diagnosis in the high phosphorous group, serum phosphorous decreased by a mean of 0.91 mg/dL (n = 33, P = .006). Finally, the relative likelihood of each assessment being completed after making the related diagnosis at the previous visit compared with those for whom that diagnosis was not made was assessed, including the likelihood of a patient's protein intake assessed after a diagnosis in the insufficient protein group was made (odds ratio = 4.08, P < .05).
CONCLUSIONS
This study demonstrates the clinical utility of a web-based HD-specific nutrition algorithm, including the ability to track changes in outcomes over time. There is potential for future research to use this tool and investigate the comparative impact of nutrition interventions.
Collapse