Table 8Clinical evidence profile: Matron/nurse-led care versus usual care

Quality assessmentNo of patientsEffectQualityImportance
No of studiesDesignRisk of biasInconsistencyIndirectnessImprecisionOther considerationsAll interventionsControlRelative (95% CI)Absolute
All-cause mortality (follow-up 6 weeks - 2 years)
34randomised trialsserious1no serious inconsistencyno serious indirectnessno serious imprecisionNone

629/3868

(16.3%)

629/3512

(17.9%)

RR 0.88 (0.8 to 0.98)21 fewer per 1000 (from 4 fewer to 36 fewer)

⨁⨁⨁◯

MODERATE

CRITICAL
Length of stay (days) (follow-up 6 weeks - 1 year; Better indicated by lower values)
12randomised trialsno serious risk of biasserious3no serious indirectnessno serious imprecisionNone11281167-MD 0.51 lower (1.33 to 0.31 lower)

⨁⨁⨁⨁

HIGH

CRITICAL
Quality of life (high score is good) - Barthel Index (follow-up 1 year; Better indicated by higher values)
1randomised trialsvery serious1no serious inconsistencyno serious indirectnessserious2None116135-MD 3.99 higher (0.97 to 7.01 higher)

⨁◯◯◯

VERY LOW

CRITICAL
Quality of life (high score is good) - QoL Myocardial Infarction Questionnaire (follow-up 100 days; Better indicated by higher values)
1randomised trialsserious1no serious inconsistencyno serious indirectnessno serious imprecisionNone134133-MD 8.40 higher (0.08 lower to 16.88 higher)

⨁⨁⨁◯

MODERATE

CRITICAL
Quality of life (high score is good) - SF-36 Physical component (follow-up 12-24 weeks; Better indicated by higher values)
2randomised trialsserious1serious3no serious indirectnessserious2None141138-MD 10.78 higher (3 lower to 24.56 higher)

⨁◯◯◯

VERY LOW

CRITICAL
Quality of life (high score is good) - SF-36 Mental component (follow-up 12-24 weeks; Better indicated by higher values)
2randomised trialsserious1serious3no serious indirectnessserious2None142142-MD 7.15 higher (0.88 lower to 15.17 higher)

⨁◯◯◯

VERY LOW

CRITICAL
Quality of life (high score is bad) (follow-up 60 days - 2 years; Better indicated by lower values)
9randomised trialsserious1serious3no serious indirectnessno serious imprecisionNone735799-MD 3.09 lower (5.43 to 0.75 lower)

⨁⨁◯◯

LOW

CRITICAL
Admission (>30 days; continuous data) (follow-up 3-12 months; Better indicated by lower values)
6randomised trialsno serious risk of biasno serious inconsistencyno serious indirectnessno serious imprecisionNone649624-MD 0.04 higher (0.06 lower to 0.14 higher)

⨁⨁⨁⨁

HIGH

CRITICAL
Admission (>30 days; dichotomous data) (follow-up 6 weeks - 2 years)
28randomised trialsserious1serious3no serious indirectnessno serious imprecisionNone

1403/3145

(44.6%)

1339/2877

(46.5%)

RR 0.90 (0.82 to 1)47 fewer per 1000 (from 84 fewer to 0 more)

⨁⨁◯◯

LOW

CRITICAL
Re-admission (follow-up 30 days - 1 year)
2randomised trialsserious1no serious inconsistencyno serious indirectnessserious2None

62/220

(28.2%)

70/220

(31.8%)

RR 0.89 (0.67 to 1.17)35 fewer per 1000 (from 105 fewer to 54 more)

⨁⨁◯◯

LOW

CRITICAL
GP visits (continuous data) (follow-up 6-12 months; Better indicated by lower values)
2randomised trialsserious1no serious inconsistencyno serious indirectnessno serious imprecisionNone171126-MD 0 higher (1.05 lower to 1.04 higher)

⨁⨁⨁◯

MODERATE

IMPORTANT
GP visits (dichotomous data) (follow-up 3-24 months)
5randomised trialsserious1serious3no serious indirectnessserious2None

332/486

(68.3%)

404/529

(76.4%)

RR 0.88 (0.75 to 1.03)92 fewer per 1000 (from 191 fewer to 23 more)

⨁◯◯◯

VERY LOW

IMPORTANT
Emergency department admissions (continuous data) (follow-up 6-12 months; Better indicated by lower values)
4randomised trialsno serious risk of biasserious3no serious indirectnessno serious imprecisionNone453420-MD 0.05 lower (0.38 lower to 0.28 higher)

⨁⨁⨁◯

MODERATE

IMPORTANT
Emergency department admissions (dichotomous data) (follow-up 4 weeks - 12 months)
8randomised trialsserious1serious3no serious indirectnessserious2None

132/531

(24.9%)

168/524

(32.1%)

RR 0.74 (0.51 to 1.06)83 fewer per 1000 (from 157 fewer to 19 more)

⨁◯◯◯

VERY LOW

IMPORTANT
Patient satisfaction (high score is good) (follow-up 60 days - 10 months; Better indicated by higher values)
2randomised trialsno serious risk of biasno serious inconsistencyno serious indirectnessno serious imprecisionNone227232-MD 1.26 higher (0.24 to 2.27 higher)

⨁⨁⨁⨁

HIGH

IMPORTANT
Patient satisfaction (high score is bad) (follow-up 30 days; Better indicated by higher values)
1randomised trialsserious1no serious inconsistencyno serious indirectnessserious2None166166-MD 0.2 lower (0.33 to 0.07 lower)

⨁⨁◯◯

LOW

IMPORTANT
Patient dissatisfaction; dichotomous data (follow-up 6 months)
1randomised trialsvery serious1no serious inconsistencyno serious indirectnessserious2None

115/223

(51.6%)

119/247

(48.2%)

RR 1.07 (0.89 to 1.28)34 more per 1000 (from 53 fewer to 135 more)

⨁◯◯◯

VERY LOW

CRITICAL
1

Downgraded by 1 increment if the majority of the evidence was at high risk of bias, and downgraded by 2 increments if the majority of the evidence was at very high risk of bias.

2

Downgraded by 1 increment if the confidence interval crossed 1 MID or by 2 increments if the confidence interval crossed both MIDs.

3

Heterogeneity, I2=50%, p=0.04, unexplained by subgroup analysis.

From: Chapter 9, Community nursing

Cover of Emergency and acute medical care in over 16s: service delivery and organisation
Emergency and acute medical care in over 16s: service delivery and organisation.
NICE Guideline, No. 94.
National Guideline Centre (UK).
Copyright © NICE 2018.

NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.