Intervention Review

You have free access to this content

Nutritional screening for improving professional practice for patient outcomes in hospital and primary care settings

  1. Amir-Houshang Omidvari1,
  2. Yasaman Vali2,3,
  3. Susan M Murray4,
  4. David Wonderling5,
  5. Arash Rashidian6,7,*

Editorial Group: Cochrane Pain, Palliative and Supportive Care Group

Published Online: 6 JUN 2013

Assessed as up-to-date: 29 MAY 2013

DOI: 10.1002/14651858.CD005539.pub2


How to Cite

Omidvari AH, Vali Y, Murray SM, Wonderling D, Rashidian A. Nutritional screening for improving professional practice for patient outcomes in hospital and primary care settings. Cochrane Database of Systematic Reviews 2013, Issue 6. Art. No.: CD005539. DOI: 10.1002/14651858.CD005539.pub2.

Author Information

  1. 1

    Tehran University of Medical Sciences, Knowledge Utilization Research Center (KURC), Tehran, Tehran, Iran

  2. 2

    Tehran University of Medical Sciences, School of Medicine, Tehran, Iran

  3. 3

    Tehran University of Medical Sciences (TUMS), Knowledge Utilization Research Center (KURC), Tehran, Iran

  4. 4

    Royal College of Physicians, London, UK

  5. 5

    Royal College of Physicians, National Clinical Guideline Centre, London, UK

  6. 6

    Tehran University of Medical Sciences, Department of Health Management and Economics, School of Public Health, Tehran, Iran

  7. 7

    Tehran University of Medical Sciences, Knowledge Utilization Research Center, Tehran, Iran

*Arash Rashidian, arashidian@tums.ac.ir. arash.rashidian@gmail.com.

Publication History

  1. Publication Status: Edited (no change to conclusions)
  2. Published Online: 6 JUN 2013

SEARCH

 
Characteristics of included studies [author-defined order]
Moore 1997

MethodsRandomised clinical trial


Participants26 community based practices of internists and family physicians in Los Angeles. 261 patients including 112 patients in intervention group (Mean age: 77, Male/Female: 0.53) and 149 patients in control group (Mean age: 76, Male/female: 0.69). Inclusion criteria: patients with age >70yrs, English speaking, not acutely or terminally ill, and able to answer questions visiting their physicians for a new visit or physical examination


InterventionsComprehensive health screening that included nutritional screening. Application of a questionnaire including malnutrition/weight loss, visual impairment, hearing loss, cognitive impairment, urinary incontinence, depression, physical limitations, and reduced leg mobility.

Physicians in the intervention group were individually briefed in their offices for 30 to 45 minutes regarding the study, the screening instrument, and the accompanying clinical 1-3 page clinical summaries. The guidelines were succinct to encourage physicians to use them during a clinical encounter. Physicians in the intervention group were given manuals containing more extensive versions of the clinical summaries and two clinically pertinent articles for each one of them.


OutcomesHealth status, nutrition problems detection rate, nutritional intervention rate

Other outcomes reported in this paper but not relevant to this review were relevant to screenings for vision, hearing, memory, incontinence, depression, physical function, leg mobility problems


Notes


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low riskThe physicians were randomised using a table of random numbers

Allocation concealment (selection bias)Low riskGiven the study design, there is a low risk of bias

Incomplete outcome data (attrition bias)
All outcomes
Unclear risk31 of 261 patients did not complete the entire study. Similar loss to follow-up of about 12% in both groups

Selective reporting (reporting bias)Low riskThe relevant outcomes in the methods section have been reported in the results section

Other biasUnclear riskStaff were not blinded and although they were trained to avoid selection bias, office staff may have asked patients to participate who were healthier and therefore more likely to be able to fill out the study consent forms and questionnaires without assistance

Blinding of participants and personnel (performance bias)
All outcomes
High riskPersonnel, clinicians and patients were not blind in the study

Blinding of outcome assessment (detection bias)
All outcomes
Unclear riskNot stated in the text

Jordan 2003

MethodsControlled before-after study


ParticipantsAll patients admitted during study periods (May 1999, January 2000), and considered by nurses to be able to give informed consent were approached for permission to review their notes and observe mealtimes. Ultimately 175 patients (mean age: 67.5, Male/Female: 0.88) with hospitalisation for 1-56 days (mean=9.1) participated in the study in four groups. 46 and 27 patients in stage 1 May 1999 in intervention and comparator ward respectively. 64 and 38 patients in stage 2 January 2000 in intervention and comparator ward respectively.


InterventionsNutritional screening. Applying a Nursing Nutritional Screening Tool which comprised a form that scores patients on a scale of 1–4 for mental condition, weight or weight loss, appetite, ability to eat, gut function, medical condition and pressure ulcers. The score is interpreted as follows; <15, high risk; 16–21, moderate risk; 22–28, minimal risk. Teaching sessions were held in November 1999 to introduce the screening tool.


OutcomesNursing documentation, patient care at mealtime, referral to dietitian


Notes


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)High risk

Allocation concealment (selection bias)High risk

Selective reporting (reporting bias)Low risk

Other biasHigh risk

Rypkema 2003

MethodsControlled trial and cost-consequence study


Participants140 geriatric ward inpatients of a university hospital (mean age= 81.3) in the intervention group and 158 inpatients of geriatric ward of a non-academic teaching hospital (mean age = 82.6) in control group were participated. All of them were non-terminally ill, older than 60yrs and admitted for more than 2 and less than 150 days during a 10-month period in 2001.


InterventionsNutritional screening. All patients in the intervention group were screened for risk of malnutrition by nurses and if positive screened using a fuller nutrition screening tool. They were also screened for dysphagia and dehydration. Patients with one positive screening test were investigated by a dietitian and a speech and language therapist and a geriatrician. Main treatments were standard high energy diet or supplement, swallowing therapy and rehydration.


OutcomesLength of stay, weight change, more than 3% weight loss, pressure sores, nosocomial infection, total hospital cost per patient


Notes


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)High riskNo randomisation was performed

Allocation concealment (selection bias)High riskNot applicable

Incomplete outcome data (attrition bias)
All outcomes
Unclear riskNot stated in the text explicitly

Selective reporting (reporting bias)Low riskThe relevant outcomes in the methods section have been reported in the results section

Other biasHigh riskParticipants were selected from two separate hospitals. This would prevent cross-contamination but might increase confounding factors effects

Blinding of participants and personnel (performance bias)
All outcomes
Unclear riskNot stated in the text explicitly

Blinding of outcome assessment (detection bias)
All outcomes
Unclear riskNot stated in the text

 
Characteristics of excluded studies [ordered by study ID]

StudyReason for exclusion

Abayomi 2004Study design not meeting the inclusion criteria

Anonymous 1994Not a primary study

Anonymous 1994bNot a primary study

Anonymous 1996Not a primary study

Anonymous 1996bNot a primary study

Anonymous 1998Not a primary study

Anonymous 1999Not a primary study

Anonymous 2000Study design not meeting the inclusion criteria

Apovian 2001Not a primary study

Azad 1999Study design not meeting the inclusion criteria

Barrocas 1995Study design not meeting the inclusion criteria

Berstein 2004Not a primary study

Bolin 2007Type of intervention

Briend 1995Not a primary study

Brown 2006Study design not meeting the inclusion criteria

Brugler 1999Study design not meeting the inclusion criteria

Campbell 1994Study design not meeting the inclusion criteria

Capra 2007Not a primary study

Cereda 2006Not a primary study

Cerede 2008Not a primary study

Christman 2009Study design not meeting the inclusion criteria

Cotton 1996Study design not meeting the inclusion criteria

Cren 2004Non-English language

Erskine 2006Study design not meeting the inclusion criteria

Ferro-luzzi 1966Not a primary study

Finn 1993Not a primary study

Garabige 2005Non-English language

Garofalo 1995Study design not meeting the inclusion criteria

George 1994Not a primary study

Ghosh 1976Study design not meeting the inclusion criteria

Goodheart 1983Not a primary study

Graves 1982Study design not meeting the inclusion criteria

Green 2002Not a primary study

Halliday 2010Not a primary study

Ham 1991Not a primary study

Hebert 2001Not a primary study

Hegland 1991Not a primary study

Holmen 2006Study design not meeting the inclusion criteria

Houda 1993Study design not meeting the inclusion criteria

Huhmann 2008Study design not meeting the inclusion criteria

Hunt 1985Study design not meeting the inclusion criteria

Jiang 2007Non-English language

Kita 1996Not a primary study

Klein 1997Study design not meeting the inclusion criteria

Kruizenga 2005Study design not meeting the inclusion criteria

Kuszajewski 2005Study design not meeting the inclusion criteria

Li 2002Non-English language

Mclaren 1998Study design not meeting the inclusion criteria

Melchior 2002Non-English language

Melnick 1994Study design not meeting the inclusion criteria

Micklewright 1999Not a primary study

Mitchell 2002Study design not meeting the inclusion criteria

Molis 1999Study design not meeting the inclusion criteria

Mosely 1981Not a primary study

Muller 2003Non-English language

Perry 2003Study design not meeting the inclusion criteria

Pironi 2000Not a primary study

Poshthauer 2006Not a primary study

Queen Sammoure 1995Not a primary study

Reilly 1996Not a primary study

Reynard 2006Non-English language

Reynish 2001Not a primary study

Rondanelli 2006Non-English language

Sayarath 1993Study design not meeting the inclusion criteria

Sutor 1998Study design not meeting the inclusion criteria

Tannen 2007Non-English language

Tramposch 1987Study design not meeting the inclusion criteria

Triano 1982Study design not meeting the inclusion criteria

Velzeboer 1983Study design not meeting the inclusion criteria

Vincent 2010Type of intervention

White 1992Study design not meeting the inclusion criteria

Wilson 1996Not a primary study

Wood 1970Study design not meeting the inclusion criteria

Yanochik 1976Study design not meeting the inclusion criteria

 
Summary of findings for the main comparison.

Study IDOutcomesBeforeAfterComments


Control group (mean)NIntervention

group (mean)
NDifference
(95% CI)
P ValueControl group (mean)NIntervention group (mean)NDifference
(95% CI)
P Value

Moore 1997General health score57112611494(-3 to 10)0.31709969131-1(-8 to 4)0.53

Nutrition problems detection rate (%)NR#511251490(-1 to 1)0.98









Nutritional intervention rate (%)NR411241490(-4 to 4)0.84










Study IDOutcomesControl group (%)NIntervention

group (%)
NControl group (%)NIntervention group (%)NDifference
(95% CI)
P Value

Jordan 2003Dietitian referrals18.52715.246-3.310.5389.464-1.1

Incidence of weight documentation29.6272646-3.67.938726464.1

Study IDOutcomesControl groupNIntervention

group
NDifference
(95% CI)
P ValueControl groupNIntervention groupNDifference
(95% CI)
P Value

Rypkema 2003Length of stay (days - mean)NR32.715831.11401.60.51









Weight change (Kg - mean)NR-0.761400.921051.68<0.001









Number of patient with >3% weight lossNR4214011105-19%<0.001









Number of nosocomial infectionsNR5815833140-13%0.01









Number of patients with pressure soresNR3315823140-0.4%0.37









Mean hospital cost per patient (Euro)*NR79081587516140-392

(-1619 to 835)

 # NR: Not Reported.
*Costs were compared by measuring time spent on training and extra time needed for screening monitoring and intervening in the intervention group; materials used, such as diagnostic tests for infection and supplements, in control and intervention group; and number of in-hospital days for both groups. No discounting has been performed, as the study duration is less than one year. No price year was mentioned.
 
Table 1. General description of included studies

IDTitleStudy designCountrySettingInterventionStudy periodParticipantsMain relevant outcomes


PreInterventionPostIntervention groupControl group

Moore 1997A Randomized Trial of Office-based Screening for common
Problems in Older Persons
Randomised Controlled TrialUSACommunity, general practiceComprehensive health problem screeningNot stated in the textpatients >70yrs, English speaking, not acutely or terminally ill,able to answer questions visiting their physicians for a new visit or physical examination1.Ferequency of nutrition problems detection and 2.Intervention as a result of detection of nutrition problems

112149










Jordan 2003Introducing a nutrition screening tool: an exploratory study in a district general hospitalQuasi experimental, Controlled before-after studyUKHospitalNutritional ScreeningMay 1999November 1999January 2000All patients admitted during study periods and able to give informed consent1.Nursing documentation, 2.patient care at mealtime and 3.referral to dietitian

May:46

Jan:27
May:64

Jan: 38

Rypkema 2003Cost-effectiveness of an interdisciplinary intervention in geriatric inpatients to prevent malnutritionProspective controlled trial plus a cost-consequence studyNetherlandsHospitalNutritional Screening2001non-terminally ill,60yr old inpatients of geriatric ward of two different hospitals who admitted for more than 2 and less than 150 days1.Lenght of stay, 2.Weight change, more than 3% 3.Weight loss, 4.Pressure sores, 5.Nosocomial infection

6.Total hospital cost per patient

140158