Systematic screening for undernutrition in hospitals: predictive factors for success

Since 2007, systematic screening for undernutrition has become a performance indicator (PI) for hospitals within the National Benchmarks on Quality of Care of the Dutch Health Care Inspectorate (HCI). Its introduction was guided by a national implementation program. The aim of this study was to evaluate the screening results from 2007 to 2010 and to identify predictive factors for achieved screening results.

All 97 Dutch hospitals were obliged to report screening results to the HCI. An additional questionnaire was developed to determine hospital characteristics, including hospital type, size, participation in implementation program, screening tool used, use of electronic records, presence of hospital-wide or ward task forces, and protocol-defined referral. Multivariate linear regression analysis was used to identify predictive factors for the obtained screening results in 2010.

The mean screening percentage increased from 51 ± 28% in 2007 (n = 75 hospitals, n = 340,000 patients) to 72 ± 17% in 2010 (n = 97; n = 1,050,000) (p < 0.01). Eighty-one hospitals returned the questionnaire. A higher screening percentage was associated with more clinical admissions (highest vs. lowest tertile: β = 14.0, 95% CI 3.9-20.5; p < 0.01; middle vs. lowest: β = 7.3, -0.8 to 15.6; p = 0.05), presence of protocol-defined referral to a dietician (β = 10.5, 2.9-18.0; p < 0.01), and use of the SNAQ screening tool (vs. MUST: β = 9.1, 1.7-16.6; p = 0.02).

Screening percentages have increased significantly since the introduction of the PI. Screening was more frequent in hospitals which have more patient admissions, protocol-defined referral to a dietician, and who use the SNAQ screening tool. This information may assist in improving Dutch screening rates and in implementation in other countries.