big data – big nursing - vfp-apsi.ch paans 24.11... · big data – big nursing ... systemic...

Dr. W. Paans Hanze University Groningen24.11.16

BIG DATA – BIG NURSING

24th November 2016 at AarauNursing Diagnoses and Length of Stay in Orthopedic Surgery

Dr. Wolter PaansDr. Maria Muller-Staub

Dr. Wim KrijnenHanze University of Applied Sciences, Groningen,

The Netherlands

1


Research Group in Nursing Diagnostics Intro.

Research topics: • Communication, Critical Reasoning, Patient

Involvement, Documentation and Handover• Digital Health• Family Care / Relation Based Care

Research seen from a holistic, systemic point of view

Dr. W. Paans Hanze University Groningen24.11.16Titel presentatie aanpassen 4

Research Question

17.01.16

What is the predictive power of nursing diagnosis documentation in the patient record on Length Of Hospital Stay (LOS)?

Hip prostheses patients; age of > 65, admitted in hospitals for surgery.

Knee prostheses patients; age of > 65, admitted in hospitals for surgery,

. 3


Method

17.01.16

Review of 300 records in hip prostheses patients.

Review of 604 records in knee prostheses patients.

Two orthopedic units, one Dutch hospital

Review: September 2014 - February 2016

Reference:

NANDA-I nursing diagnoses

4


Sample

Review of 300 patient records in hip prostheses patients: mean (SD) age: 76 (11) 220 female, 80 male. Review of 604 patient records in knee prostheses patients: mean (SD) age: 69 (8) 413 female, 191 male.


Data collection & instrument

17.01.16

•Measurement:

first day Post-Surgery (PS)

•Instrument:

D-Catch instrument developed for the analysis of the accuracy in nursing documentation.

Reference: NANDA-I, NIC, NOC

1Paans, W., Sermeus, W., Nieweg, M.B., Schans, van der, C.P. (2010). Psychometric properties of the D-Catch instrument, an instrument for evaluation of the nursing documentation in the patient record, Journal of Advanced Nursing,

Nursing, 66 (6), 1388-1400.

1Paans, W., Sermeus, W., Nieweg, M.B., Schans, van der, C.P. (2010). Prevalence and accuracy of nursing documentation in the patient record. Journal of Advanced Nursing, 66 (11), 2481-2490, published on line: doi:10.1111/j.

1365-2648.2010.05433.x

6


Nursing Diagnoses & L.O.S. In Hip Patients!!!

!

Nursing!diagnosis!!!!!!!!!!!!!!!!!!!!(n=!300!records)!!

%!(n)! Mean!(SD)!L.O.S.!!

!!!!P-value* !

!!!!!!!!!!!!!!!Diagnosed! Not!diagnosed!

Pain! 70!(210)! 10,92!(6.589)! 7,35!(3,207)! <0,000!

Disordered!/ !Distressed!! 42!(126)! 11,42!(7,564)! 9,07!(4,382)! !!0,008!

Pressure!ulcer! !!18!!(55)! 14,72!(8,833)! 8,96!(4,594)! <0,000!

Obstipation! 20!(60)! 12,73!(7,958)! 9,46!(5,428)! !!0,011!

Anxiety! 15!(45)! 12,23!(7,585)! 9,77!(5,828)! !!0,018!

Imbalanced!Nutrition!/ less!than!body!requirements!

14!(42)! 14,23!(9,615)! 9,46!(5,074)! !!0,011!

Imbalanced!fluid!volume!/ !deficient!fluid!volume!

12!(37)! 15,57!(10,265)! 9,32!(4,779)! !!0,004!

Impaired!tissue!perfusion!

!

Total!NDx!N=!613!/300!rec.!

Median!discharge!on!9th!day!including!day!of!admission/discharge!!

!

A!Independent!Samples!T-test!B!L.O.S!* !P!=!<!0,05!!

13!(38)!

!

15,34!(9,382)! 8,99!(4,451)! <0,000!

!

!

Dr. W. Paans Hanze University Groningen24.11.16Titel presentatie aanpassen 917.01.16

Nursing Diagnoses & L.O.S. In Knee Patients

8

Nursing!Diagnoses!(n!=!604!records)!

%!(n)!!!

Mean!(SD)!L.O.S!!Diagnosed!!!!!!!!!!!!!!!!!!!Not!Diagnosed!

P-value* !

Impaired!tissue!perfusion! 16,0!!!!(97)! 5.44!!!(3.31)! 4.85!(3.12)! <0,000!!

Disoriented!/ !Distressed! 3,80!!!!(23)! 6,48!!!(2,57)! 4,88!(3,16)! <0,000!!

Nausia! 18,15!!(110)! 6,75!!!(4,93)! 4,54!(2,42)! <0,000!!

Anxiaty! 3,30!!!!(20)! 5,35!!!(2,71)! 4,93!(3,16)! !!!0,001!!

Pain!in!rest!situation!!Delirium!!Pressure!Ulcer!!Obstipation!!Diarrhea!!Total!NDx:!N=!551/604!rec.!!Median!discharge!on!4th!day!(Including!day!of!admission!&!discharge)!!

!

31,35!!(190)!!1,16!!!!(7)!!2,31!!!!(14)!!9,74!!!!(59)!!5,12!!!!(31)!

6,54!!!(3,01)!!9,71!!!(2,81)!!10,07!(5,03)!!7,59!!!(2,78)!!9,61!!!(4,45)!

4,67!(3,18)!!4,89!(3,11)!!4,82!(3,00)!!4,65!!(3,05)!!5,73!!(3,30)!!!!!!!!!!!!!!

!!!0,011!!!!!!!!0,029!!!<0,000!!!<0,000!!!<0,000!!!!!!!

! ! !A!Independent!Samples!T-test!B!L.O.S!* !P!=!<!0,05!!!!!


Pain measurements and L.O.S. In Knee patients

Pain!scores!VNRS!/ !VAS!!(n!=!604!records)!!

%!(n)!!

Mean!(SD)!!

P-value!

0-3! 60,89!(369)! 4,76!(3,38)! 0,030* !4-7! 21,12!(128)! 5,27!(3,09)!8-10! 2,31!!!(14)! 5,71!(2,55)!!Missing!Values!

!15,68!(93)!

!

!

A!Kruskal-Wallis!H!test!B!L.O.S.!* !P!=!<!0,05!


Medical treatment & L.O.S. Knee Patients

Treatment!(n!=!604)!

%!(n)!!

Mean!(SD)!L.O.S.!!!!

P-value!!

Rapid!Recovery!! 45,00!(270)! 4,80!(1,93)!!

0,000* !

Joint!Care! 24,00!(116)! 6,33!(2,41)!!

Regular!treatment!!!Missing!values!!!n=38!

31,00!(180)! 7,90!(5,20)!!

!

!

A!Kruskal-Wallis!H!test!B!L.O.S!* !P!=!<!0,05!


Medical treatment: Knee Prostheses

Protheses!!(n!=!606)!

%!(n)!!!

Mean!(SD)!L.O.S.!!! P-value!!

Total!Knee!prostheses! 68,32!(414)! 4,62!(2,44)! 0,000* !!Total!Knee!prostheses!with!

Patella!prostheses!!12,87!(78)!

!5,90!(2,74)!

Demi!Knee!prostheses! 2,15!!!(13)! 2,54!(0,97)!Patella!prostheses! 8,75!!!(53)! 4,51!(1,97)!Revision!Total!Knee!prostheses! 5,45!!!(33)! 8,36!(7,69)!Revision!Totale!Knee!with!Patella!Prostheses!

!0,66!!!(4)!

!8,75!(6,95)!

Revision!Patella!prostheses! 1,49!!!(9)! 3,78!(1,41)!Missing!Values! 0,33!!!(2)! ! !!

A!Kruskal-Wallis!H!test!B!L.O.S!* !P!=!<!0,05!!


Medical treatment & L.O.S. Hip Patients

!

Medical!treatment!(n=300)!

!

%!(n)! !!!!!!Mean!(SD)!L.O.S!!

!P-value* !

!!0,051!

! !

Dynamic!hip!screw!(DHS)! 18!(47)! 11,11!(7,403)!

Cannula!hip!screws! 11!(29)! 8,59!(3,647)!

Gamma!nail! 49!(128)! 10,39!(6,786)!

Hemi!arthroplasty!(hip!prosthesis)!

12!(32)! 9,22!(4,680)! !

Other!treatments!!

Missing!values!n=!38!A!Kruskal-Wallis!H!test!B!L.O.S!* !P!=!<!0,05!!

10!(26)! -! -! -!

!

!


Readmissions and L.O.S. In Knee patients

Readmissions!(n!=!604)!

%!!(n)! Mean!(SD)!LOS!! P-value!

Readmitted! 5,12!!!!(31)! 6,23!(4,57)! 0,001* !!!

Not!Readmitted! 94,88!(573)! 4,87!(3,04)!

A!Independent!Samples!T-test!B!L.O.S.!* !P!=!<!0,05!


Titel presentatie aanpassen 11

Difference in Length of Stay (LOS) hip patients and medical diagnoses

ᴬ Independent samples T-testᴮ Dependent variable: LOS, p<.05.

19-01-1614

!

!

!

!

!

Medical!diagnoses!(Cases:!n=!300)!

%!(n)!

!

Mean!(SD)!L.O.S.!!!

P-value!

Diagnosed! Not!diagnosed!

Lung!disease! 18!(55)! 11,31!(7,643)! 9,90!(5,778)! 0,0457* !

Cadiac!disease! !!41!(123)! 11,00!(7,152)! 9,11!(4,475)! 0,0027* !

CVA!(Stroke)! !7!(21)! 14,20!(8,170)! 9,88!(5,948)! 0,0405* !

Diabetes!

Co!morbidity!Medical!Dx!N=!275!/ !300!

!

16!(47)! 12,03(8,241)! 9,80!(5,635)! 0,0430* !


Titel presentatie aanpassen 11

Difference in Length of Stay (LOS) knee replacement and medical diagnoses

ᴬ Independent samples T-testᴮ Dependent variable: LOS, p<.05.

19-01-1615

A!Independent!Samples!T-test!

!

Comorbidity!/ !medical!Dx.!N=!433!/604!!!

B!L.O.S.!* !P!=!<!0,05!T-test!

Comorbidity!!(n!=!604)! %!(n)!!

Mean!(SD)!LOS!!!Diagnosed!!!!!!!!!!!!!!!!!!Not!Diagnosed!

P-value!!

Diabetes! 16!(99)! 5,54!(5,03)! 4,83!(2,62)! 0,011* !Hart!failure! 26!(155)! 5,90!(4,65)! 4,61!(2,35)! 0,000* !Lung!disease!! 15!(95)! 4,73!(5,03)! 4,80!(2,64)! 0,003* !Tractus!Digestivus!!disease! 8!!!(46)! 6,30!(3,79)! 4,83!(3,!07)! 0,016* !Thrombosis! 6!!!(38)! 6,82!(6,88)! 4,82!(2,69)! 0,001* !

Dr. W. Paans Hanze University Groningen24.11.16Titel presentatie aanpassen 1419-01-16

Results from modeling days hospitalized by Poisson regression in terms of the estimated parameters, their standard errors (SE), t-value, significance measured by

p-value, the rate ratio and their 95% Confidence interval.

Hip sample

16

!

Estimate! SE! t.value! P-value!

Exp!

Estimate! CLL! CLR!

(Intercept)! 1,2688! 0,3657! 3,47! 0.0000! 3,5567! 1,7312! 7,2595!

Age! 0,0103! 0,0043! 2,3788! 0,0181! 1,0104! 1,0018! 1,019!

Impaired!tissue!perfusion!!

(surgical!wound!area)! 0,3423! 0,0768! 4,46! 0,0000! 1,4082! 1,2091! 1,6338!

Pressure!ulcer! 0,2607! 0,0808! 3,2261! 0,0014! 1,2979! 1,1059! 1,5183!

Deficient!fluid!volume! 0,3464! 0,0899! 3,8546! 0,0000! 1,414! 1,1828! 1,6826!

Diabetes! 0,214! 0,0672! 3,1848! 0,0016! 1,2386! 1,0843! 1,4111!

Dr. W. Paans Hanze University Groningen24.11.16Titel presentatie aanpassen 1419-01-16

Results from modeling days hospitalized by Poisson regression in terms of the estimated parameters, their standard errors (SE), t-value, significance measured by

p-value, the rate ratio and their 95% Confidence interval.

Knee sample

17

!

!

!

! Estimate! SE! T-Value! P-Value! Exp!Estimate!

CLL! CLR!

(Intercept)! 1,2892! 0,0268! 45,1260! <!0.0000! 3,6299! 3,4307! 3,8373!!Medical!Treatment!!

!0,3755!

!0,0625!

!6,0056!

!<!0,0000!

!1,4557!

!1,2867!

!1,6440!

Pressure!Ulcer!!

0,4024! 0,1025! 3,9261! 0,0001! 1,4954! 1,2172! 1,8196!

Thrombosis!!

0,2783! 0,0693! 4,0140! 0,0001! 1,3209! 1,1503! 1,5098!

Inpaired!Tissue!Perf.!!

0,1502! 0,0510! 2,9475! 0,0032! 1,1621! 1,0505! 1,2828!

Nausea!!

0,2393! 0,0460! 5,1986! <!0,0000! 1,2704! 1,1601! 1,3896!

Delirium!!

0,6232! 0,1389! 4,4860! <!0,0000! 1,8649! 1,4055! 2,4251!

Obstipation! 0,2704! 0,0559! 4,8389! <!0,0000! 1,3105! 1,732! 1,4606!


Comparing diagnostics in the hip & knee sample

• Nursing diagnoses and comorbidity are more prevalent in hip patients compared to knee patients

• Pain is the most prevalent nursing diagnosis in both groups

• Nursing Diagnoses Impaired Tissue Perfusion and Pressure Ulcer are strong predictors of L.O.S. in both groups

• Medical treatment in knee patients is a strong predictor of L.O.S. (< L.O.S = Rapid Recovery)

• Medical treatment in hip patients is not a significant predictor of L.O.S.

• Nursing Diagnoses: Pain, Impaired Tissue Perfusion, Disordered / Distressed, Pressure Ulcer, Obstipation and Anxiety are significantly related to increased L.O.S. in both groups

• Nursing Diagnosis Nausia is a strong predictor of L.O.S. in the knee group, and significantly related to medical treatment (rapid recovery v.s. other treatments: P value <0,000)

• The nursing diagnosis Deficient Fluid Volume is a strong predictor of L.O.S. in the group of hip patients

• Thrombosis (med. diagnosis) can be seen as a risk factor in the knee patient group (prev. 38 / 604, and strong predictive on L.O.S)


Conclusions

17.01.16

Nursing interventions are documented with low accuracy; the effect on

outcomes is (still) not to measurable.

Relationship between nursing diagnoses and nursing actions /

interventions, as well as the effect of nursing interaction is hard

to measure as the nature of the documentation is descriptive

and not systematically (sometimes diffuse / cryptic, unclear and

redundant in nature)

19.12.15 19


Conclusions

17.01.16

Diagnostic information:

T1 (diagnostic assessment information): poor,

T2 (diagnostic post surgery information): moderate,

T3 (diagnostic discharge / hand over information) poor.

20


Needs for Big Data Computing

17.01.16

Technical improvements in the EHR are needed, i.e. output calculation possibilities:

- Nursing Process Decision Support Systems (NPDSS) Implementation of the use of definitions and classificatons

- Nanda-I, NIC, NOC for accuracy and efficiency in documentation

- Trans-sectorial care cooperation developments

- E-Health / interoperability to foster data exchange

- The use of new technologies (QS).

21


Recommendations for clinical practice

17.01.16

Nursing Process - Clinical Decision Support Systems (NP-CDSS) are

needed.

Nursing Process-Decision Support allows to retrieve Standardized

Nursing Data from Electronic Health Records such as nursing diagnoses

and hospital duration (LOS).

22


Measurement links

17.01.1623

Linkages of sensor techniques and nursing diagnoses in the PES structure (SSEP-I-O): signs

detections by sensor second skin applications as a validation of nurses’ observation)


RELATION BASED CARE

Essentials:

• Involvement of the patient and relatives is essential (info. accuracy)

• Critical reasoning skills are essential

• Communication skills are essential

• Documentation skills are essential

• Classification is essential

• DDSS’ are essential

An holistic approach in relation based nursing is essential


Related publication (free tekst)

An Internationally Consented Standard for Nursing Process-

Clinical Decision Support Systems in Electronic Health Records.

Müller-Staub M, de Graaf-Waar H, Paans W.

Comput Inform Nurs. 2016 Nov;34(11):493-502.

PMID: 27414705

Similar articles

https://www.ncbi.nlm.nih.gov/pubmed/27414705

https://www.ncbi.nlm.nih.gov/pubmed?linkname=pubmed_pubmed&from_uid=27414705


To know more about the professorship in Nursing Diagnoses?

Google or You Tube: type in:

Modern Times in Nursing

https://www.youtube.com/watch?v=1TcmOtCBz54




Literature

17.01.16

• Alameda, C., Suárez, C.(2009) Clinical outcomes in medical outliers admitted to hospital with heart failure. European Journal of Internal Medicine 20 764–767.

• Belcher, J.V.R., Alexy, B. (1999) High-Resource Hospital Users in an Integrated Delivery System. The Journal of Nursing Administration.Vol 29(10), p 30-36.

• Cots, F., Mercadé , L., Castells, X. Salvador, X.(2004) Relationship between hospital structural level and length of stay outliers Implications for hospital payment

systems. Health Policy 68 159–168.

• Freitas, A., Silva-Costa, T., Lopes, F., Garcia-Lema, I., Teixeira-Pinto, A., Brazdil, P., and Costa-Pereira, A. (2012). Factors influencing hospita; high length of stay

outliers. BMC Health Services Research.

• Hauskrecht, M., Valko, M., Batal, I., Clermont, G., Visweswaran, S. & Cooper, G.F. (2010) Conditional Outlier Detection for Clinical Alerting. AMIA Annu Symp Proc.

p. 286–290.

• Kuwabara, K., Imanaka, Y., Matsuda, S., Fushimi, K., Hashimoto, H., Ishikawa, K.B., Horiguchi, H., Hayashida, K., Fujimori, K.(2008) The association of the number

of comorbidities and complications with length of stay, hospital mortality and LOS high outlier, based on administrative data. Environ Health Prev Med 13:130–137

• Müller-Staub, M., Lavin, M.A., Needham, I. & Achterberg, T van (2006). Nursing diagnoses interventions and outcomes – application and impact on nursing

practice: systematic review. Journal of Advanced Nursing, 56, 514-531

• Paans, W., Sermeus, W., Nieweg, R., Schans, van der, C.P. (20102). Prevalence and accuracy of nursing documentation in the patient record.Journal of Advanced

Nursing, 66 (11), 2481-2490, published on line: doi:10.1111/j.1365-2648.2010.05433.x

• Perimal-Lewis L., Hakendorf Li J.Y., Ben-Tovim D.I., Qin S., Thompson C.H. (2012) The relationship between in-hospital location and outcomes of care in patients

of a large general medical service. Internal Medicine Journa l- Royal Australasian College of Physicians.

• Veer, A.J.E de & Francke, A.L. (2009). Attitudes of nursing staff towards electronic patient records: a questionnaire survey. International Journal of Nursing Studies.

• Welton J.M., & Halloran E.J. (2005). Nursing diagnoses, diagnosis-related group, and hospital outcomes. JONA

• Xiao, J., Douglas, D., Lee, A.H. and Vemuri, S.R.A (1997). Delphi evaluation of the factors influencing lenght of stay in Australian hospitals. International Journal of

Health planning and management. Vol 12, 207-218.

27

http://www.ncbi.nlm.nih.gov/pubmed?term=Perimal-Lewis%20L%5BAuthor%5D&cauthor=true&cauthor_uid=23279255

http://www.ncbi.nlm.nih.gov/pubmed?term=Li%20JY%5BAuthor%5D&cauthor=true&cauthor_uid=23279255

http://www.ncbi.nlm.nih.gov/pubmed?term=Ben-Tovim%20DI%5BAuthor%5D&cauthor=true&cauthor_uid=23279255

http://www.ncbi.nlm.nih.gov/pubmed?term=Qin%20S%5BAuthor%5D&cauthor=true&cauthor_uid=23279255

http://www.ncbi.nlm.nih.gov/pubmed?term=Thompson%20CH%5BAuthor%5D&cauthor=true&cauthor_uid=23279255

big data – big nursing - vfp-apsi.ch paans 24.11... · big data – big nursing ... systemic...

Documents