Initial Implementation in Western Uganda J. Lucian Davis, MD, MAS - - PowerPoint PPT Presentation

The Severe Illness Management System (SIMS) Platform: Initial Implementation in Western Uganda

• J. Lucian Davis, MD, MAS

Epidemiology of Microbial Diseases Pulmonary, Critical Care, & Sleep Medicine Lucian.Davis@yale.edu 15 December 2016

Disclosures

• I have no conflicts of interest to disclose.

2

Roadmap

• Global implementation gap in severe illness care
• Development of the SIMS platform in Uganda
• SIMS 1.0 study findings

3

What is severe illness?

• Life-threatening syndromes encountered in hospitals

–In-hospital mortality ≥ 10% –Common physiologic responses to diverse diseases

• Maternal & child illnesses
• Injuries from man-made or natural disasters
• Infectious diseases - HIV/AIDS & other emerging pandemics
• Non-communicable diseases

–A leading cause of death & disability in young adults

4

What are severe illness conditions?

• Coma

– Failure of central nervous system

• Severe respiratory distress

– Failure of oxygenation or ventilation

• Shock

– Failure of circulatory system

• Sepsis

– Dysregulated host response to infection

5

Why focus on severe illness care?

• Preventable deaths can be avoided if clinicians have

the tools to provide high-quality severe illness care

• Front line clinicians are uniquely positioned to detect,

report, and contain emerging public health threats presenting as undifferentiated severe illness.

6

What is the scope of the problem?

7

Africa (n=263) High-income countries (n=44) Capable of implementing all Grade 1 recommendations for sepsis? 5.7% 91%

“Walimu aims to save the lives of severely ill patients in low-income countries by enhancing the quality of hospital care.”

www.walimu.org

What can be done for severe illness?

www.walimu.org/imai

IMAI: Integrated Management of Adult & Adolescent Illness

9

Quick Check Severe Illness Algorithm

10

www.walimu.org/imai

13 of 26

WHO IMAI Quick Check Training

11

Defining Target Behaviors: Process Map

12

Measure Vital Signs & Perform Physical Exam Diagnose Severe Illness Treat Severe Illness

Link to a Theory of Behavior Change

13

Michie et al Implement Sci 2011 Michie et al Implement Sci 2011

Measure Vital Signs & Perform Physical Exam Diagnose Severe Illness Treat Severe Illness

Formative assessment

14

COM-B: Barriers to severe illness care

15

Capability Opportunity Motivation Deliver high-quality care for severe illness

Knowledge of severe illness care Skills to resuscitate severely ill Time & staff to provide severe illness care Supplies & equipment to deliver severe illness care Social influences allowing change Professional identity to deliver quality Belief in capability to change practice Belief in capability to change outcomes Intention to bring about change

Michie et al Implement Sci 2011

SR: Health worker performance

16

Rowe AK et al. USAID Seminar. 31 March 2015

BCW: Tailoring interventions to barriers

17

Capability Opportunity Motivation

Supportive supervision & Collaborative improvement meetings Quick Check Training & mGuidelines Clinical Mentoring Audit & Feedback plus

Michie et al Implement Sci 2011

Severe Illness Management Support (SIMS) Platform

Deliver high-quality care for severe illness

TIDieR: Specifying the SIMS intervention

18

SIMS Components Clinical Mentoring Supportive Supervision & Collaborative Improvement Meetings Audit & Feedback Why Knowledge & skills gap Environmental context & resources gap Reinforcement gap What Teaching rounds Problem solving Monitoring & reinforcement Who provides Visiting expert clinician Local clinician champion Semi-automated How Shadowing at the bedside In person Email & SMS Where At the hospital At the hospital At the hospital When & how much All day every 4 months Bi-monthly for one hour Bi-monthly (Report) Weekly (SMS) Tailoring Education, Training, Modeling Environmental restructuring Enablement, Persuasion Modifications Add distance mentoring? Added process improvement fund Review in collaborative improvement meetings How well Twice About monthly at 3 sites As designed

Hoffman T et al BMJ 2014

SIMS 1.o Sites

19

Kasese District

Bwera Hospital Kagando Hospital Kilembe Mines Hospital

• St. Paul’s Health Centre

SIMS 1.0 Study Design

20

Quasi-randomized, stepped-wedge design

Kagando Kilembe

• St. Paul

Bwera Months Aug 2014 Sep Oct Nov Dec Jan 2015 Feb Mar Apr May Baseline Intervention Endline

5,759 patients admitted 1,633 Pre-Intervention Cohort 4,126 Intervention Cohort

Enrollment

Hospital 1 Hospital 2 Hospital 3 Hospital 4 759 (46.5) 117 (7.2) 663 (40.6) 94 (5.8) 1336 (32.4) 1018 (24.7) 1516 (36.7) 256 (6.2)

21

Patient characteristics

22

Characteristic n (%) Pre-intervention period n=1633 Intervention period n=4126 Women 59% 57% Age, median years 38 (24-55) 37 (23-58) HIV-seropositive 20% 14% Admitting diagnosis Malaria 34% 37% Pneumonia 5% 4% Heart failure 4% 3% Urinary tract infection 4% 5% Length of stay, median days (IQR) 3 (2-6) 3 (2-6)

Impact of SIMS on Vital Sign Collection

23

Vital sign checked? (n=5759) Pre- (%) Post- (%) Change (%) 95% CI P-value Temperature 27 49 +22 +18 to +26 <0.001 Heart rate 10 32 +22 +11 to +32 <0.001 Pulse oximetry 19 +19 +18 to +20 <0.001 Blood pressure 56 69 +13 +9 to +16 <0.001 Respiratory rate 5 10 +5 +1 to +9 0.008 Mental status 11 15 +4 +1 to +7 0.004 HIV status 37 45 +8

• 8 to +24

0.33

Inter-site variation: Pattern 1

24

Temperature Blood pressure

Inter-site variation: Pattern 2

25

Respiratory Rate Mental Status Assessment

Impact on Severe Illness Diagnosis

26

Severe illness? (n=5759) Pre- (%) Post- (%) Risk Ratio 95% CI P-value Shock 11 17 1.53 0.9 – 2.5 0.090 Sepsis 0.4 4 10.1 2.3 – 31 <0.001 Respiratory distress 1 4 5 2.4 – 3.8 <0.001 Altered mental status 5 4 0.69 0.6 – 0.8 <0.001

In-hospital Mortality

• Trend towards lower mortality in intervention period?

–4.3% vs 3.7%, -0.6%, 95% CI -2.3 to +1.1, p=0.48

• Presence of severe illness strongly predicted mortality

–Risk Ratio 2.6, 95% CI 2.4-2.7, p<0.001

27

Conclusions

• SIMS, a theory-informed intervention to improve

health worker performance, was feasible & effective in a low-income country:

– Improved vital sign collection – Increase in severe illness diagnoses – No definitive effects on treatment quality or mortality

• Significant heterogeneity by site/vital sign/condition
• Ongoing work on fidelity and adaptation in order to

refine interventions / implementation strategy for future replication and scale-up

28

Acknowledgements

29

Matt Cummings, MD Shevin Jacob, MD, MPH Achilles Katamba, MBChB, PhD Olive Kabajaasi, B.A. Savio Mwaka, B.A. Nathan Kenya-Mugisha, MBChB Elijah Goldberg, B.A. Adithya Cattamanchi, MD, MAS Funders World Health Organization US Defense Threat Reduction Agency Anonymous European Family Foundation D43TW009607 (JLD)

And especially our patients and clinicians

30

Extra Slides

31

Inter-site variation: Pattern 3

32

Heart Rate Oxygen Saturation

Inter-site variation: Pattern 4

33

HIV Status Assessment

Mobile guidelines

34