Patients are doing it for themselves The Evidence for Self-administration of Home IV Therapy

Pauline Dobson & Dr Mark Loewenthal

Immunology & Infectious Diseases Unit

John Hunter Hospital, Newcastle NSW

and

Faculty of Health, University of Newcastle

2 of 47

• 40 years ago the first accounts of home IV

antibiotic courses replacing inpatient stays

were reported – the first HITH.1

– The antibiotics were self-administered

• In 2013, the bulk of this treatment is still

clinician administered

• Recently evidence has been published to

support self-administration as a model of

care

1. Antoniskis A, Anderson BC, Van Volkinburg EJ, et al: Feasibility of outpatient self-

administration of parenteral antibiotics. West J Med 128:203-206, Mar 1978

3 of 47

• Today’s Nurses

– Highly educated workforce

– IV therapy commonplace

– Skilled with Central venous catheters

– Better knowledge about how to avoid infection

– Clear guidelines and policies

• Surely it follows that nurses would

be much better than patients or carers self-administering

4 of 47 Image from: Quarksandquirks.wordpress.com

However we may be making assumptions…

5 of 47

Examples of some

clinician managed

PICCs in patients

referred to our

service from

elsewhere

6 of 47

Unsecured

Multiple

Opsite

dressings

Excess

External

length

7 of 47

St Elsewhere’s a number of days post insertion

Unsecured,

bloody,

two

dressings

8 of 47

The long

and

winding

road

9 of 47

Policy “…the dressing is changed on a weekly basis or immediately if the

integrity of the dressing is compromised.”

10 of 47

Phlebitis

Multiple

Opsites

Multiple

Opsites

Multiple

Opsites

Multiple

Opsites

Blood,

Statlock

Placement

Old spot

bandage

under Opsite

11 of 47

Grotty

Double Bunger

12 of 47

Neat

Triple Bunger

Neat

Triple Bunger

Neat

Triple Bunger!

13 of 47

14 of 47

• 1993-2005, 2059 admissions over 13 years; 473 episodes

(24%) self administration

• Catheter complications overall 1.5/1000 catheter days (lit)

• Readmission Clinician 12.6%, Self 10.5% (no significant

difference)

15 of 47

• 2001-2011, 2766 episodes, 42,238 days, 69% PIVCs &

butterfly devices, 23% midlines, 5% tunnelled TCVC, 1.5%

PICCs

• For analysis 854 midline, PICC, & TCVC for 25,292 days

• Line infection 0.8 / 1000 cds

• Flucloxacillin use OR 3.0 for other line events (given q.i.d)

16 of 47

• Prospective cohort, paediatric & adult patients

admitted to Out & About from 1/10/1995 –

31/12/12

• Only home parenteral antibiotic therapy included

• Clinician administration

– community nurse, RACF nurse, HITH nurse in clinic

• Self-administration

– either patient or their carer administers IV therapy

17 of 47

• Out & About Home IV Therapy Program

• Home IV therapy, with parenteral

antibiotics as majority of admissions

• Infectious Diseases led

• Commenced in 1995

• Paediatric and adult patients - co-located

children's hospital

• Clinical database recording outcomes

18 of 47

~3hrs

198km

HITH Service Geographic boundaries

~6hrs

427km

19 of 47

• Offered as option only

• Patient / Carer must be willing to self-administer

• Assessed for: – Cognition

– Hygiene

– Motor skills (dexterity, tremor)

– Compliance

– IDU history

– Ability to read written instructions (language)

– Hearing (alarms, telephone contact)

– Vision (pump controls)

20 of 47

• Teaching session

• One usually adequate

• May require more for CADD pumps,

or multiple antibiotics

• Extension set, to allow patients to flush PICC using

both hands

• Every patient, regardless of whether they are self

or clinician administration, are contacted daily by

phone

• Check temp, PICC / pump status, complications, falls

21 of 47

• Early discharge

– The patient does not complete the intended course of

treatment due to a complication

• Readmission

– The patient is readmitted to hospital for more than 48

hours following a deterioration or new condition

• On call use

– The patient contacts the HITH team to resolve a

problem, either in or after hours

• Call Out

– The patient requires HITH staff to meet the patient to

troubleshoot their problem e.g. blocked catheter

22 of 47

• Bone & Joint Infection

– Osteomyelitis, septic arthritis, prosthetic joint infection, orthopaedic

hardware (pins, plates, screws, rods) infection, discitis & bursitis

• Abscess

– liver/splenic; spinal/epidural; cerebral; lung; psoas; & deep tissue

• Cellulitis is serious or limb threatening

• Multiple other includes:

– malignant otitis media; encephalitis; ascending cholangitis;

empyema; nocardia; meningitis; parotitis; leptospirosis; pyomyositis;

pericarditis

– infected devices e.g. pacemakers, permacaths, CVCs, cochlear

implants, VP shunts, peritoneal dialysis catheters;

23 of 47

Variable Clinician Administered Self-Administered Combined

Admissions 2739 1913 4652

Patients 2204 1091 3295

Patient days 59,983 42,920 102,903

Median Length of stay 21 20 21

Gender – Male 1758 (64.2%) 1107 (57.9%) 2865 (61.6%)

Age – (mean, range) 58.8 (3 months - 96yrs) 40.0 (3 weeks – 92yrs) 51.1

Paediatric 140 377 477

Females are less likely to utilise self-

administration, Odds Ratio 0.65 (95% CI 0.47

to 0.89). Note: self is often carer.

24 of 47

• 2705 (82%) patients

admitted once only

• 391 (12%) had two

admissions

• 199 patients (6%) from

3 – 33 admissions

• Small numbers of

patients have repeated

admissions

Admissions

Per patient Number of

Patients Percent

1 2,705 82.1 2 391 11.9 3 91 2.76 4 35 1.06 5 14 0.42 6 12 0.36 7 5 0.15 8 4 0.12 9 6 0.18 10 5 0.15 11 5 0.15 12 5 0.15 13 2 0.06 14 2 0.06 15 3 0.09 18 1 0.03 19 3 0.09 20 1 0.03 21 1 0.03 22 1 0.03 27 2 0.06 33 1 0.03

Total 3,295 100

25 of 47

Variable Clinician Self Combined

Bone & Joint infection 1552 (57%) 854 (45%) 2406 (51.7%)

Cystic Fibrosis 81 (3%) 497 (26%) 578 (12.4%)

Infective Endocarditis 208 (8%) 61 (3%) 269 (5.8%)

Abscess 168 (6%) 83 (4%) 251 (5.4%)

Sepsis/Bacteraemia 128 (5%) 83 (4%) 211 (4.5%)

Wound Infection 138 (5%) 48 (3%) 186 (4%)

Bronchiectasis 88 (3%) 75 (4%) 163 (3.5%)

Cellulitis 92 (3%) 42 (2%) 134 (2.9%)

Other 284 (10%) 170 (9%) 454 (9.8%)

26 of 47

Vascular Device Clinician Self Combined

PICC 2510 (88.6%) 1516 (76.8%) 4026 (83.7%)

Implantable Port 73 (2.6%) 364 (18.4%) 437 (9%)

CVC 197 (6.9%) 68 (3.4%) 265 (5.5%)

Tunnelled Catheter 29 (1%) 22 (1.1%) 51 (1%)

Peripheral IVC 25 (0.9%) 3 (0.15%) 28 (0.6%)

Total 2,834 1973 4,807

Note: more lines than

admissions, some lines need

replacing during admission

27 of 47

Drug Clinician Self Total

Flucloxacillin 835 518 1,353

Vancomycin 503 214 717

Benzylpenicillin 398 183 581

Timentin 290 237 527

Meropenem 91 193 284

Ceftazidime 76 198 274

Tobramycin 51 218 269

Cephalothin 192 46 238

Tazocin 30 103 133

Ceftriaxone 88 40 128

Gentamicin 57 58 115

Teicoplanin 57 50 107

Cephazolin 55 23 78

Cefepime 16 48 64

Piperacillin 5 33 38

Total 2744 2162 4,906

Note: more drugs than

admissions, some patients

on dual or triple tx

28 of 47

Mode Clinician Self Total

Continuous 24h 2,599 (90%) 1,684 (75%) 4,283

Intermittent infusion 98 (3%) 119 (5%) 217

Bolus 190 (7%) 435 (20%) 625

Total 2,887 2,238 5,125

29 of 47

Clinician Administered

Self Administered

30 of 47

• The proportion of those

who self-administer

has steadily increased

• And remains the case

when Cystic Fibrosis is

excluded

50%

40%

31 of 47

There was no difference in early discharge due a

complication between Self and Clinician. Age, line type,

and cystic fibrosis did not predict early discharge.

32 of 47

Female patients were more likely to be discharged early from HITH than males. On

average early discharge occurred at a rate of 9.7 per 1000 patients days for females and

7.1 per 1000 patient days for males (HR 1.40 95%CI 1.17 to 1.68 P = 0.0002)

33 of 47

34 of 47

Predictor Haz. Ratio P-value [95% Conf.Interval]

PICC 1 reference

CVC .82 0.138 0.64 to 1.1 Port .73 0.042 0.54 to .99

Tunneled Catheter .65 0.120 0.37 to.1.12

Child (<18) 1.4 <0.0001 1.2 to 1.8 Female 1.3 <0.0001 1.1 to 1.4

Cystic fibrosis .79 0.195 0.55 to 1.1 Female with CF .58 0.015 0.37 to 0.90

35 of 47

Predictor Haz. Ratio P-value [95% Conf. to Interval]

CF 0.466 <0.0001 0.31 to 0.69

child 1.69 0.002 1.2 to 2.3

female 1.33 0.003 1.1 to 1.6

There were 447

admissions resulting in

at least one after-hours

call-out. 278 (10.6%) of

Clinician patients and

169 (9.3%) self

administered

(P-value 0.26 PPP test.)

36 of 47

Administration Lines Line Days Failures Rate per 1000 line

days

Clinician 2721 57445 225 3.9

Self 1896 40693 125 3.1

Total 4617 98138 350 3.6

37 of 47

Clinician lines

have a slightly

higher

cumulative

hazard at all

times

38 of 47

Predictor Hazard Ratio P-value [95% Conf. Interval]

Self administering 0.684 0.001 0.54 - 0.86

PICC 1 - reference

CVC 1.67 0.003 1.2- 2.4

Implanted Port 0.0467 <0.001 0.011 - 0.19

Tunneled Catheter 0.445 0.168 0.14 - 1.4

Age (per year) 0.98 <0.0001 0.975 - 0.986

Each variable above is independent of the other

The important predictors of line failure are:

(1) Clinician-administration do worse after controlling for age and catheter

(2) Compared to PICCs; CVC are worse and ports are better

(3) Age: the older the better

39 of 47

Hazard Ratio P-value [95% Conf. Interval]

Self 0.273 <0.0001 0.14 to 0.53

Age (per year) 0.975 <0.0001 0.96 to 0.99

40 of 47

Lines Line Days Events Rate per 1000 line days

Clinician PICC 2427 51874 23 0.443

Self PICC 1447 32770 8 0.244

Clinician CVC 193 3715 2 0.538

Self CVC 65 1499 2 1.33

Total 4132 89858 35 0.390

There was a trend

toward less thrombosis

in the Self group after

adjustment for the

finding that thrombosis

is more likely to occur in

children but the small

numbers made the

results imprecise (P-

value 0.086 PPP test).

41 of 47

17 PICCs, 15

CVCs, and 1

tunneled line

were accidently

removed by

patients or fell

out. 27 of these

were in the

Clinician group (P <0.024 PPP test).

It can happen at

any time in the life

of the line.

42 of 47

There were only 8 definite infections in 98,528 line days

of observation. There was no significant difference

between Self & Clinician (P=0.12 PPP test)

43 of 47

Device Line Days Rate per 1000 line days

PICC 84611 0.31

CVC 5214 1.1

Total 89836 0.089

No

significant

difference

between Self

& Clinician

(P=0.43 PPP

test)

44 of 47

Factor Haz. Ratio P-value [95% Conf. Interval]

Self 1.31 0.275 0.81 to 2.1

PICC 1 reference

CVC 0.767 0.654 0.24 to 2.4

Port 0.661 0.428 0.24 to 1.9

Age <18 3.85 <0.0001 2.0 to 7.5

72 lines needed

removal for blockage.

No significant

difference between

Self and Clinician (P =

0.39, PPP test). The

only significant risk

factor was age under

18 years. May be due

to smaller lumen size

45 of 47

• Self-administration survey of Out & About Patients

– Advantages

• “not tied down”; “control over situation” “family life normal”, “get back to work”, “helps understanding of disease”

– Choose self-administration again: 93%

• Service Advantages

– Suitable for rural areas where limited numbers of nursing staff working 7 days / week & evenings

– Capacity of service is not as affected by number of HITH team staff

– Costs lower: fewer staff, cars & transport

• Service Disadvantages

– Requires thorough assessment prior to acceptance

46 of 47

• In selected patients, self-administration is

safe, and equivalent or better in

outcomes to clinician administered home

parenteral antibiotic therapy

• Gender and age were important factors in

HITH success, independent of who

administered the HPAT

47 of 47

Questions?

Corresponding author: pauline.dobson@hnehealth.nsw.gov.au