A Study of HIV Presentations to an Adult Emergency Department and Development and Validation of an Outcome Prediction Model

March 3, 2020Uncategorized

CANDIDATE’S
SURNAME: Laher FIRST
NAME/S: STUDENT NUMBER:
CURRENT QUALIFICATIONS: MBBCh, MMed, FCEM, Cert Critical Care, EDIC, DipPEC, DCH, DipAllerg
TEL: CELL: E-MAIL: FAX: N/A
DEGREE FOR WHICH PROTOCOL IS BEING SUBMITTED: PhD
PART-TIME OR FULL-TIME: Full time
FIRST REGISTERED FOR THIS DEGREE: TERM : April YEAR: 2016
DEPARTMENT: Department of Emergency Medicine
TITLE OF PROPOSED RESEARCH: A Study of HIV Presentations to an Adult Emergency Department and Development and Validation of an Outcome Prediction Model
CANDIDATE’S SIGNATURE: DATE: 25 April 2016
SUPERVISOR’S NAME: Professor % Supervision: 33.4%
SUPERVISOR’S QUALIFICATIONS: MBBCh, PhD, FCP, FRCP, FCCP
SUPERVISOR’S DEPARTMENT: HOD, Critical Care, Charlotte Maxeke Johannesburg Academic Hospital
SUPERVISOR’S ADDRESS / TEL / E-MAIL: Cell: / E-mail:
SUPERVISOR’S NAME: Professor Fathima Paruk % Supervision: 33.3%
SUPERVISOR’S QUALIFICATIONS: MBChB, MD, FCOG, Cert Critical Care
SUPERVISOR’S DEPARTMENT: HOD, Critical Care,
SUPERVISOR’S ADDRESS / TEL / E-MAIL: Cell: / E-mail:
SUPERVISOR’S NAME: % Supervision: 33.3%
SUPERVISOR’S QUALIFICATIONS: MBBCh, MMed, FCP, Cert ID, DTM&H, Dip HIV Man
SUPERVISOR’S DEPARTMENT: Deputy Executive Director, Wits Reproductive Health and HIV Institute
SUPERVISOR’S ADDRESS / TEL / E-MAIL: Cell: / E-mail:
SYNOPSIS OF RESEARCH:
Sub-Saharan Africa has the highest prevalence of HIV in the world, housing more than 2/3 of global cases of HIV. South Africa has more people with HIV than any other country in the world. The 2014 UNAIDS HIV and AIDS estimates for South Africa, estimated 18.9% of South African adults over the age of 15 years as being HIV positive. Over the last few years, access to anti-retroviral therapy (ART) in South Africa has significantly improved. The most recent (2015), National Department of Health consolidated guidelines for the management of HIV in South Africa have significantly simplified and relaxed the criteria for the initiation of ART’s in order to ensure that the correct therapy is started at the right time. Despite this our emergency departments (ED) and hospital wards continue to bear a high burden of HIV associated morbidity and mortality. The aims of this PhD study are: 1) to describe the clinical presentation, pathology and outcomes of adult HIV positive patients presenting to the CMJAH ED, 2) to determine antiretroviral therapy non-adherence rates and factors contributing to poor adherence in adult HIV positive patients presenting to the CMJAH ED, 3) to develop an outcome prediction model for adult HIV positive patients presenting to the CMJAH ED, 4) to internally validate the outcome prediction model that will be developed. I will hope to achieve this via four concurrent studies, each concentrating on one of the aims.
ETHICS APPROVED: Y IF  SUPPLY ETHICS CLEARANCE No: M160512
SIGNATURE OF SUPERVISOR/S:
SIGNATURE PG OFFICE STAFF

………………………………… REGISTERED

YES….. NO…..
STAMP

University of the Witwatersrand
Faculty of Health Sciences
School of Medicine

A Study of HIV Presentations to an Adult Emergency Department and Development and Validation of an Outcome Prediction Model

Degree: PhD

Student Number:

Supervisors:
Professor
Professor
Professor

April 2016

TABLE OF CONTENTS
LITERATURE REVIEW AND MOTIVATION TO CONDUCT THIS STUDY 1
METHODOLOGY 3
STUDY 1: CLINICAL PRESENTATION, PATHOLOGY AND OUTCOMES OF ADULT HIV POSITIVE PATIENTS PRESENTING TO THE CMJAH ED 3
Aim 3
Objectives 3
Study Design 3
Site of Study 3
Study Population 4
Inclusion Criteria 4
Exclusion Criteria 4
Sample size 4
Data Collection 5
Data Analysis 6
Limitations 6
STUDY 2: DETERMINING SOCIAL AND COMPLIANCE RELATED REASONS FOR HEALTH DETERIORATION OF ADULT HIV POSITIVE PATIENTS PRESENTING TO THE CMJAH ED 6
Aim 6
Objectives 6
Study Design 6
Site of Study 7
Study Population 7
Inclusion Criteria 7
Exclusion Criteria 7
Sample size 7
Data Collection 7
Data Analysis 7
Limitations 8
STUDY 3: DEVELOPMENT OF AN OUTCOME PREDICTION MODEL FOR ADULT HIV POSITIVE PATIENTS PRESENTING TO THE CMJAH ED 8
Aim 8
Objectives 8
Study Design 8
Site of Study 8
Study Population 8
Inclusion Criteria 8
Exclusion Criteria 8
Sample size 8
Data Collection 9
Data Analysis 9
Limitations 9
STUDY 4: VALIDATION OF AN OUTCOME PREDICTION MODEL FOR HIV POSITIVE PATIENTS PRESENTING TO THE CMJAH ED 9
Aim 9
Objectives 9
Study Design 10
Site of Study 10
Study Population 10
Inclusion Criteria 10
Exclusion Criteria 10
Sample size 10
Data Collection 10
Data Analysis 10
Limitations 10
ETHICAL CONSIDERATIONS 11
CONSENT 11
BUDGET 11
FUNDING 11
TIMEFRAME 11
POST DOCTARAL WORK 12
DISSEMINATION OF FINDINGS 12
REFERENCES 12
APPENDICES 14
Appendix A: CMJAH ED HIV Study Data Collection Sheet 14
Appendix B: Request for permission from hospital clinical director and H.O.D to conduct research study 17
Appendix C: Patient information sheet 20
Appendix D: Deferred patient information sheet (unable to get next of kin / legal guardian consent) 22
Appendix E: Deferred patient information sheet (with next of kin / legal guardian consent) 24
Appendix F: Next of kin / legal guardian information sheet 26
Appendix G: Doctor / medical staff information sheet 28
Appendix H: Consent form (patient) 30
Appendix I: Consent form (patient – deferred) 32
Appendix J: Consent form (next of kin / legal guardian) 34
Appendix K: Ethics clearance certificate 36

LITERATURE REVIEW AND MOTIVATION TO CONDUCT THIS STUDY
Sub-Saharan Africa has the highest prevalence of HIV in the world, housing more than two-thirds of global cases of HIV. South Africa has an estimated 6.4 million people living with HIV, i.e. more people with HIV than any other country in the world.1 The 2014 UNAIDS HIV and AIDS estimates for South Africa, estimated 18.9% of South African adults over the age of 15 years as HIV positive.2

Anti-retroviral therapy (ART) has been available in the public sector since 2004. Over the last decade, access to ART’s in South Africa has become increasingly easier.3 Currently there are more than 2.6 million South Africans receiving ART’s that are now freely available at > 3500 health facilities throughout SA.4,5 In an attempt to ensure that the correct therapy is started at the appropriate time, the most recent (2015) National Department of Health consolidated guidelines pertaining to the management of HIV in South Africa have significantly simplified and relaxed the indications for the initiation of ART. As such ART’s are now recommended for all pregnant and breastfeeding women as well as all children less than 5 years of age regardless of the CD4 cell count. ART’s are also recommended for all individuals over 5 years of age with WHO stage 3 or 4 disease as well as in any individual who has a CD4 cell count ≤500 cells/μl, irrespective of the WHO stage of disease.6

Although the life expectancy of individuals initiated on ART’s in SA has reportedly increased to approximately 80% of normal life expectancy,7 our emergency departments (ED) and hospitals continue to encounter a high burden of HIV positive patients. The associated morbidity and mortality as well as the cost and resource allocation implications are substantive in this patient population.8 A study conducted recently in the medical wards of a Cape Town district hospital, revealed that 35.7% of HIV positive patients were unaware of their HIV status at the time of their admission. In this study the most frequent reasons for admission constituted newly diagnosed TB (33.5%), other bacterial infections (17.1%) and AIDS defining illness (10.9%). The authors reported a 13.3% HIV related mortality. Independent predictors of mortality were AIDS-defining illnesses other than TB, anaemia and renal dysfunction. HIV accounted for almost two-thirds of the medical admissions.9

The reasons for health deterioration in patients with HIV include the development of drug resistance, poor compliance / adherence, denial and drug-related toxicities. This in turn may lead to virological and immunological failure with clinical progression of disease.10 Common reasons for ART non-compliance include: poor education (lack of knowledge related to the disease and its treatment), forgetting to take the medication, unavailability of stock at the dispensing site, occupation, large number of tablets which are considered to be too much to consume, work related travel, side effect intolerance, lack of food, financial crisis, depression and poor social support.11,12

Determining outcomes in medical patients is not an easy task. Outcome prediction in medicine may be either subjective or objective. Although subjective (‘gut feel”) outcome prediction is widely used in daily practice, it is prone to human error.13 Objective outcome prediction models provide significant adjunctive information for clinicians. Validated outcome prediction scoring tools provide clinicians with additional information that may be useful when counselling patients and their families. This may potentially improve patient / family satisfaction and reduce the risk of litigation. Other advantages may include; the selection of appropriate therapeutic strategies, better allocation of resources, timely initiation of palliative care, appropriate patient recruitment in research trials and comparison of unit performance.14 Various outcome prediction models have been developed and are in common use.15–17

Afessa and Green conducted an outcome prediction study consisting of 141 ICU patients with HIV disease. They reported a poor correlation between CD4 counts and in-hospital mortality. In this study, 12% of all hospitalised patients were admitted to ICU. The APACHE II score predicted mortality was 29.6%, whereas the actual mortality was 45.2%.18 Although higher APACHE II scores were associated with poorer outcomes, the large disparity in this study suggests that current outcome prediction models may not be accurate and appropriate in HIV positive patients. This is not surprising, as the development of APACHE II did not take in to consideration HIV disease (in terms of severity). What this highlights is that a prediction tool that factor in the severity of HIV disease is required for this population.

At the Charlotte Maxeke Johannesburg Academic Hospital Emergency Department (CMJAH ED), which is the proposed study site, approximately 120 HIV positive patients present every month (as per ED unit patient register), many of whom have already been initiated on ART’s. The epidemiology of HIV and underlying reasons for health deterioration in this population is not known. A literature search using common biomedical databases such as PubMed, Scopus, Web of Science and Google Scholar was conducted. Keywords such as emergency department, accident and emergency, casualty, HIV, AIDS, antiretroviral therapy (ART) failure, ART resistance, outcome prediction, validation and prognosis did not produce valid results for outcome prediction models in the HIV population presentating to the Emergency Department.

This PhD study will aim to:
1) Describe the clinical presentation, pathology and outcomes of adult HIV positive patients presenting to the Charlotte Maxeke Johannesburg Academic Hospital Emergency Department.
2) Determine antiretroviral therapy non-adherence rates and factors contributing to poor adherence in adult HIV positive patients presenting to the CMJAH ED.
3) Develop an outcome prediction model for adult HIV positive patients presenting to the Charlotte Maxeke Johannesburg Academic Hospital Emergency Department.
4) Internally validate the outcome prediction model that will be developed.

The above will be achieved by conducting four concurrent studies addressing each of the aforementioned objectives.

METHODOLOGY
This PhD project will comprise 4 studies. Data for all four studies will be collected simultaneously and concurrently using a single data collection sheet (appendix A). The studies are entitled as follows:
Study 1: Clinical presentation, pathology and outcomes of adult HIV positive patients presenting to the CMJAH ED
Study 2: Antiretroviral therapy non-adherence rates and factors contributing to poor adherence in adult HIV positive patients presenting to the CMJAH ED
Study 3: Development of an outcome prediction model for adult HIV positive patients presenting to the CMJAH ED
Study 4: Validation of an outcome prediction model for HIV positive patients presenting to the CMJAH ED

STUDY 1: CLINICAL PRESENTATION, PATHOLOGY AND OUTCOMES OF ADULT HIV POSITIVE PATIENTS PRESENTING TO THE CMJAH ED

Aim
To describe the clinical presentation, pathology and outcomes of adult HIV positive patients presenting to the CMJAH ED

Objectives
To describe the demographics of adult HIV positive patients presenting to the CMJAH ED
To describe outcomes including mortality and final disposition of adult HIV positive patients presenting to the CMJAH ED
To describe the current clinical presentation including history, examination and investigation findings of adult HIV positive patients presenting to the CMJAH ED
To describe the final diagnoses of adult HIV positive patients presenting to the CMJAH ED

Study Design
Single-centre, prospective, cross-sectional, observational, descriptive study

Site of Study
The study will be conducted at the Charlotte Maxeke Johannesburg Academic Hospital Emergency Department (CMJAH ED) situated in Parktown, Johannesburg, South Africa. CMJAH ED is a quaternary level academic facility that manages all non-trauma, priority 1 and 2 patients (priority classified as per the South African Triage Score19), older than 18 years that have come through the ED triage system. Priority 3 patients and stable patients that are not from the drainage area of CMJAH are referred away from the triage area to the appropriate facility.

Study Population
This study will aim to include the first 800 consecutive patients with underlying HIV that have been triaged in to the CMJAH ED (area 165 &167) from the date of approval of this study. As part of routine practice, all patients that have been triaged into the ED with an unknown HIV status are routinely counselled to undergo HIV testing.

Inclusion Criteria
Consecutive patients older than 18 years presenting to the CMJAH ED with underlying HIV

Exclusion Criteria
Patient refusal of consent or unable to obtain consent to participate in the study

Sample size
The sample size calculation for this study is based on the sum of the number of subjects that will be required to develop the outcome prediction model in study 3 and the number of subjects that will be utilized to validate the prediction model in study 4.

The sample size calculation for study 3 was derived as follows:
Assuming that mortality is 14% and we want to estimate this to within 5% precision at a significance level of 5%, the minimum sample size required is 186 patients, using the formula:
n=(Z^2 P(1-P))/d^2
Where:
n = sample size,
Z = Z-statistic for the chosen level of confidence,
P = expected prevalence or proportion
d = precision
If 10% do not consent, the required sample size would be 204

A sample size of 50% of the sample size required for study 3 will be needed for the validation phase of study 4. Therefore study 4 will require at least 102 subjects (204/2). Hence study 1 should comprise at least 306 (204 +102) subjects.

However, an attempt will be made to collect data from at least 1200 subjects in the entire study. Studies 1 will analyse data of all 1200 subjects. Study 2 will include all subjects with poor adherence to ART’s, which is estimated to be between 150-200 subjects. Study 3 will utilise data of the first 600 (50% of entire sample) subjects to develop the outcome prediction model. Study 4 will utilise data of the latter 600 (50% of entire sample) subjects to attempt to validate the developed model.

As per the CMJAH ED patient register, an average of 4 patients with confirmed HIV are seen daily. Therefore it is estimated that data collection will be completed within 12-18 months from the date of initiation of data collection.

Data Collection
Once permission has been granted to commence the study, all doctors working at the CMJAH ED will be individually informed regarding the study design and requirements (appendix E).
Doctors working on each shift will be asked to inform the primary investigator (Dr A Laher) regarding attendance of patients with confirmed HIV and patients that have consented to HIV testing.
Patients with an unknown HIV status that had consented to HIV testing will be followed up by the primary investigator. Those with a positive result will also be approached to participate in the study.
The process of obtaining informed consent and data collection will be conducted by the primary investigator.
Informed consent will be obtained simultaneously for all four studies (appendices C – J).
In the event that the potential subject is not in the position to grant consent (e.g. decreased level of consciousness), consent will be obtained from the next of kin / legal guardian. Once the patient is capacitated to consent then informed consent will be obtained from the patient. If the patient refuses consent, their data will be destroyed and they will be excluded from the study.
Data relevant to the study will be obtained from the patients’ hospital record.
Relevant findings not found in the patient’s hospital record will be directly obtained from the patient or next of kin / legal guardian where applicable. If required, the patient may need to be examined again by the primary investigator.
Only if the next of kin / legal guardian is already aware that the patient is HIV positive, will they be questioned with regards to history directly related to HIV (e.g. treatment regimen, treatment compliance etc.).
As per CMJAH ED unit policy, the following laboratory investigations are routinely performed for all patients with newly diagnosed or known HIV presenting to the unit: 1) CD4+ cell count, 2) HIV viral load, 3) CMV viral load, 4) arterial blood gas (ABG), 5) urea and electrolytes (U+E), 6) Calcium, magnesium, phosphate and albumin (CMP) and 6) CRP. The following investigations are only performed when indicated; 1) βD Glucan 2) liver function test (LFT), 3) INR/PTT, 4) lumbar puncture, 5) sputum analysis, 6) Bone marrow analysis and 7) radiological investigations.
Relevant data will be captured into the data collection sheet (Appendix A).
To complete data collection, patients’ hospital records will be followed up daily for the entire duration of hospital stay (for patients that will be admitted).
Data will then be transferred to an electronic data capturing system, REDCap ©.
It will be ensured that the data collection process will not interfere with the unit and researcher’s clinical duties and responsibilities.

Data Analysis
Specific data from the data collection sheet pertinent to this leg of the study will be analysed using STATA 14, (StataCorp. 2015. Stata Statistical Software: Release 14. College Station, TX: StataCorp LP). A descriptive analysis of the data will be conducted. Categorical data will be described by frequency and percentage tabulation, and will be illustrated by pie and bar charts. Continuous data will be described by the mean, standard deviation, median, and interquartile range, and the distribution of the data will be analysed by histograms. Kaplan-Meier survival plots will be used to describe mortality.

Limitations
It is likely that a proportion of patients that will not meet triage criteria to be treated at CMJAH will be HIV positive. Some patients refusing consent for HIV testing may also be HIV positive. The inability to obtain a complete history may also be another limitation to the study as patients / escorts may not be unaware or unwilling to divulge certain aspects of the history.

STUDY 2: Antiretroviral therapy non-adherence rates and factors contributing to poor adherence in adult HIV positive patients presenting to the CMJAH ED

Aim
To determine underlying social and compliance related reasons for health deterioration of adult HIV positive patients presenting to the CMJAH ED

Objectives
To determine the rates of ART non-adherence in adult HIV positive patients presenting to the CMJAH ED
To determine and describe the underlying factors and reasons contributing to poor treatment adherence with anti-retroviral therapy (e.g. denial, unavailability, side effects, poor education) in adult HIV positive patients presenting to the CMJAH ED
To determine the correlation between a detectable HIV viral load and the 4 item in the Adult AIDS Clinical Trial Group (AACTG) adherence score.in adult HIV positive patients presenting to the CMJAH ED

Study Design
Single-centre, prospective, cross-sectional, observational, descriptive study

Site of Study
As for study 1 (above)

Study Population
As for study 1 (above)

Inclusion Criteria
As for study 1 (above)

Exclusion Criteria
As for study 1 (above)

Sample size
The rates of ART non-adherence is not known in acutely ill HIV positive patients presenting to the CMJAH ED, however based on expert opinion, we are estimating a non-adherence rate of approximately 20%, which would equate to between 150-200 subjects in this study.

Data Collection
Consecutive subjects enrolled in study 1 that were previously ever initiated on ART’s and now presenting with virolological failure as evidenced by a detectable HIV viral load on laboratory testing will be approached for more detailed questioning regarding ART adherence (appendix 2).
As for study 1, in the event that the potential subject is not in the position to grant consent (e.g. decreased level of consciousness) then consent and history will be obtained from the next of kin / legal guardian. Once the patient is capacitated to consent then informed consent will be obtained from the patient. If the patient refuses consent, their data will be destroyed and they will be excluded from the study.
Only if the next of kin / legal guardian is already aware that the patient is HIV positive, will they be questioned with regards ART adherence.
The rest of the data collection procedure will be as for study 1 (above).

Data Analysis
This is as for study 1 (above), however data specific to this arm of the study will be analysed. Spearman nonparametric correlation will be employed to investigate the associations between each of the 4 adherence items with HIV viral load and the CD4+ cell count. The receiver operating curve (ROC) will be used to compare the 4 adherence items with detectable (versus undetectable) HIV viral loads to determine the predictive validity of adherence items.

Limitations
As for study 1 (above)

STUDY 3: DEVELOPMENT OF AN OUTCOME PREDICTION MODEL FOR ADULT HIV POSITIVE PATIENTS PRESENTING TO THE CMJAH ED

Aim
To develop an outcome prediction model for adult HIV positive patients presenting to the CMJAH ED

Objectives
To develop a scoring model that will reliably determine critical outcomes including the probability of death, transfer to ICU, hospital stay > 10 days, transfer to a dependent care facility, transfer to a lower level care facility, requiring endotracheal intubation and mechanical ventilation and requiring cardiopulmonary resuscitation.

Study Design
Single-centre, prospective, cross-sectional, detection and discovery study

Site of Study
As for study 1 (above)

Study Population
The first 50% of patients recruited for study 1 will be used to develop the outcome prediction model

Inclusion Criteria
As for study 1 (above)

Exclusion Criteria
As for study 1 (above)

Sample size
The sample size calculation for this study was derived as follows:
Assuming that the risk of mortality is 14% and we want to estimate this to within 10% precision at a significance level of 5%, the minimum sample size required is 186 patients, using the formula
n=(Z^2 P(1-P))/d^2
Where:
n = sample size,
Z = Z-statistic for the chosen level of confidence,
P = expected prevalence or proportion
d = precision
If 10% do not consent, the required sample size would be 204.

However, since I intend recruiting at least 1200 patients for the entire project, the first 600 (50%) patients will be included in this study (see study 1 above).

Data Collection
As for study 1 (above)

Data Analysis
Data specific to this arm of the study will also be analysed using STATA 14, (StataCorp. 2015. Stata Statistical Software: Release 14. College Station, TX: StataCorp LP). Multiple logistic regression analysis will be performed on variables and parameter pertinent to this study. The following variables will be analysed for possible inclusion in the model: 1) ART adherence, 2) presence of mental illness, 3) drug abuse, 4) blood pressure on presentation, 5) respiratory rate on presentation, 6) oxygen saturation on presentation, 7) final diagnosis, 8) number of organ systems involved, 9) requiring endotracheal intubation / mechanical ventilation, 10) CD4+ count, HIV viral load, 11) CMV viral load, 12) lactate, 13) creatinine, 14) haemoglobin, 15) bilirubin, 16) albumin and 17) CRP, opportunistic pathogen identified.

Once the association, correlation and relationship of each of these variables with each of the following outcomes of interest a) in-hospital mortality, b) requiring ICU admission c) hospital stay > 10 days, d) transfer to a dependent care facility, e) transfer to a lower level care facility and f) requiring endotracheal intubation and mechanical ventilation is assessed, they will be considered for possible inclusion in the model.

Limitations
As for study 1 (above)

STUDY 4: VALIDATION OF AN OUTCOME PREDICTION MODEL FOR HIV POSITIVE PATIENTS PRESENTING TO THE CMJAH ED

Aim
To internally validate the outcome prediction model developed in study 3

Objectives
To internally validate the outcome prediction model developed in study 3 for its ability to predict mortality critical outcomes including the probability of death, transfer to ICU, hospital stay > 10 days, transfer to a dependent care facility, transfer to a lower level care facility and requiring endotracheal intubation and mechanical ventilation.

Study Design
Single-centre, prospective, cross-sectional, validation study

Site of Study
As for study 1 (above)

Study Population
The latter 50% of subjects recruited for study 1 will be used to validate the outcome prediction score that was developed in study 3

Inclusion Criteria
As for study 1 (above)

Exclusion Criteria
As for study 1 (above)

Sample size
The latter 600 (50%) subjects from the total of 1200 subjects will be included in this leg of the study (see study 1 above).

Data Collection
As for study 1 (above)

Data Analysis
Data specific to this arm of the study will also be analysed using STATA 14, (StataCorp. 2015. Stata Statistical Software: Release 14. College Station, TX: StataCorp LP). Split-sample and or the bootstrap technique will be used to validate the model developed in study 3. Calibration and discrimination of the model will be carried out to assess performance.

Limitations
As for study 1 (above)

THE FOLLOWING ASPECTS PERTAIN TO ALL 4 STUDIES:

ETHICAL CONSIDERATIONS
Data collection sheets will be anonymous. Patient identifying information will not be included on the data collection sheet. All data will be treated confidentially and stored in a password protected personal computer. Hard copies of the data collection sheets will be kept under lock and key. Ethics clearance has been obtained from the Human Research Ethics Committee (Medical), of the University of Witwatersrand, certificate no. M160512 (Appendix K). The Head of the Emergency Department and the Clinical Director at CMJAH have approved the study (Appendix B). At no point will the HIV status of any patient be disclosed to any family member or escort of a patient who is not aware of the patients HIV status. Only if the family member / escort are already aware that the patient is HIV positive, will the family member / escort be questioned with regards to history directly related to HIV (e.g. treatment compliance). HIV will not be mentioned in the next of kin / patient representative information sheet. National Department of Health guidelines regarding disclosure of HIV status will be strictly adhered to at all times.

CONSENT
All subjects will be informed of the requirements involved with participation, and written informed consent will be obtained prior to participation. Should consent not be possible due to the patient’s condition at the time of presentation, patients will be enrolled only once consent is obtained from the next of kin. The patient will later be approached, if and when the patient is able to give consent.

BUDGET
Expenses: Predicted Cost:
Printing R 8 000.00
Stationary R 3 000.00
Telephone R 3 500.00
Total R 14 500.00

FUNDING
This project will be funded by the CMJAH ED departmental research budget.

TIMEFRAME
May 2016 – Submission for protocol assessment and ethical clearance
August 2016 – Protocol review
September 2016 to March 2018 – Data collection
April 2018 to June 2018 – Analysis of data
July 2018 to November 2018 – Preparation of thesis and submission of articles to journals
December 2018 – Submission of thesis for marking

POST DOCTARAL WORK
External validation in collaboration with interested investigators in other settings

DISSEMINATION OF FINDINGS
Results of the study will be grouped and presented as a scientific paper. The aim is to have the results written up as scientific articles to submit for publication in international peer-reviewed journals. The research findings will also be made available to the University of the Witwatersrand.

REFERENCES
1. Shisan O, Rehle T, Simbayi L., Zuma K, Jooste S, Zungu N, et al. South African National HIV Prevalence, Incidence and Behaviour Survey, 2012. Cape Town: HSRC Press; 2014.
2. UNAIDS [Internet]. 2014 [cited 2016 Mar 21]. Available from: http://www.unaids.org/en/regionscountries/countries/southafrica
3. Abdool Karim SS, Churchyard GJ, Karim QA, Lawn SD. HIV infection and tuberculosis in South Africa: an urgent need to escalate the public health response. Lancet (London, England). 2009 Sep 12;374(9693):921–33.
4. Bekker L-G, Venter F, Cohen K, Goemare E, Van Cutsem G, Boulle A, et al. Provision of antiretroviral therapy in South Africa: the nuts and bolts. Antivir Ther. 2014 Jan;19 Suppl 3:105–16.
5. Bor J, Herbst AJ, Newell M-L, Bärnighausen T. Increases in adult life expectancy in rural South Africa: valuing the scale-up of HIV treatment. Science. 2013 Feb 22;339(6122):961–5.
6. National consolidated guidelines for the prevention of mother-to-child transmission of HIV (PMTCT) and the management of HIV in children, adolescents and adults [Internet]. Pretoria; 2015 [cited 2016 Apr 22]. Available from: http://www.sahivsoc.org/upload/documents/ART Guidelines 15052015.pdf
7. Johnson LF, Mossong J, Dorrington RE, Schomaker M, Hoffmann CJ, Keiser O, et al. Life expectancies of South African adults starting antiretroviral treatment: collaborative analysis of cohort studies. PLoS Med. 2013 Jan;10(4):e1001418.
8. Meyer-Rath G, Miners A, Santos AC, Variava E, Venter WDF. Cost and resource use of patients on antiretroviral therapy in the urban and semiurban public sectors of South Africa. J Acquir Immune Defic Syndr. 2012 Nov 1;61(3):e25–32.
9. Meintjes G, Kerkhoff AD, Burton R, Schutz C, Boulle A, Van Wyk G, et al. HIV-Related Medical Admissions to a South African District Hospital Remain Frequent Despite Effective Antiretroviral Therapy Scale-Up. Medicine (Baltimore). 2015 Dec;94(50):e2269.
10. Simoni JM, Frick PA, Pantalone DW, Turner BJ. Antiretroviral adherence interventions: a review of current literature and ongoing studies. Top HIV Med. Jan;11(6):185–98.
11. Essomba EN, Adiogo D, Koum DCK, Amang B, Lehman LG, Coppieters Y. [Factors associated with non-adherence of adults infected with HIV on antiretroviral therapy in a referral hospital in Douala]. Pan Afr Med J. 2015 Jan;20:412.
12. Oku AO, Owoaje ET, Ige OK, Oyo-Ita A. Prevalence and determinants of adherence to HAART amongst PLHIV in a tertiary health facility in south-south Nigeria. BMC Infect Dis. 2013 Jan;13:401.
13. Kassirer JP, Gorry GA. Clinical problem solving: a behavioral analysis. Ann Intern Med [Internet]. 1978 Aug [cited 2016 Apr 25];89(2):245–55. Available from: http://www.ncbi.nlm.nih.gov/pubmed/677593
14. Paruk F. Critically ill obstetric and gynaecology patients: The development and validation of an outcome prediction model. University of KwaZulu-Natal; 2007.
15. Knaus WA, Draper EA, Wagner DP, Zimmerman JE. APACHE II: a severity of disease classification system. Crit Care Med. 1985 Oct;13(10):818–29.
16. Le Gall JR, Loirat P, Alperovitch A, Glaser P, Granthil C, Mathieu D, et al. A simplified acute physiology score for ICU patients. Crit Care Med. 1984 Nov;12(11):975–7.
17. Marshall JC, Cook DJ, Christou N V, Bernard GR, Sprung CL, Sibbald WJ. Multiple organ dysfunction score: a reliable descriptor of a complex clinical outcome. Crit Care Med. 1995 Oct;23(10):1638–52.
18. Afessa B, Green B. Clinical course, prognostic factors, and outcome prediction for HIV patients in the ICU. The PIP (Pulmonary complications, ICU support, and prognostic factors in hospitalized patients with HIV) study. Chest. 2000 Jul;118(1):138–45.
19. Bruijns SR, Wallis LA, Burch VC. A prospective evaluation of the Cape triage score in the emergency department of an urban public hospital in South Africa. Emerg Med J. 2008 Jul;25(7):398–402.

APPENDICES
Appendix A: CMJAH ED HIV Study Data Collection Sheet

Subject Number: ________________
Patient demographics
Age: Gender: M F Nationality: SA Other:
Marital status: Single Married Divorced Widowed Separated
Race: Black White Indian Coloured Other:
Residence: House Flat Communal Other:
Employment: Y N What: Grant: Y N Type:
How many meals a day do you eat? 1 2 3 Other:
How often do you eat meat/fish/chicken? 1x/wk 2x/wk 3x/wk 4x/wk 5x/wk Everyday
HLOE: Referring institute:
History
Main complaint:

TB history (date, systems involved, completed treatment):

History of previous HIV related pathology admission (date, system involved, therapy received):

Allergies:
Social Hx:
Past Medical Hx:

Past Surgical Hx

Medications:

HIV history
First diagnosed (date): 1st initiated on ART’s (date):
Last CD4+ count Last CD4 date:
Last VL Last VL date:
Current ART regimen:
Test done: Rapid Elisa Refused
Previous ART regimens and reason for change: Known resistance to any ART’s to date (and reason):

Concurrent use of other therapies: Herbal Traditional Alternative medicine Other
ART adherence (if applicable)
How many pills have been prescribed daily and how many times per day?
Adherence to antiretroviral medication Yes No
Did patient miss pills yesterday? Yes No
Did patient miss pills day before yesterday Yes No
Did patient miss pills last weekend (Saturday and/or Sunday) Yes No
Did patient miss pills within the last two weeks
When did patient stop taking pills?
Contributing factors and reasons for not taking medication (mark all appropriate)
Forgetfulness Travelling / away from home
Too busy to take medication Had a change in daily routine
Couldn’t take at prescribed time Took ill/sick
Can’t tolerate side effects Thought medications were toxic/harmful
Had too many pills to take Did not want others to see me
Felt depressed Mental illness
Felt better so decided to stop Denial
Ignorance / did not know Financial crisis (no transport money)
Lack of social support Instructed to stop by Dr/traditional healer
Local clinic out of ART’s Local clinic refused to provide ART’s
Use of alcohol or any other illicit drugs such as nyaope, cocaine, marijuana (dagga) or anything else? (specify)
Please explain above and specify any other reasons:

Examination
General: Jaundice Pallor Clubbing Cyanosis Oedema Dehydration Curly eyelashes
Lymphadenopathy (extent and size):
HR: BP: RR: Sats: Temp HGT:
Respiratory:

Cardiovascular

Abdomen / Genitourinary / Gynae:

CNS / Psychiatric:

Musculo-Skeletal / Dermatology / Other:

Laboratory findings on admission (where indicated)
Blood:
CD4 count HIV-1 viral load CMV viral load
ABG pH FiO2 PaO2 PaCO2
BE Gluc Lactate Osmolality
U+E Na K Cl HCO3 Urea Cr GFR
FBC WCC Hb Plt
LFT Bili(tot) Bili(conj) Prot Alb ALP GGT
Ammonia Amylase Lipase AST ALT
Coag INR PTT(c) PTT(p) CMP Ca Mg P
Inflam CRP PCT βD Glucan
LP Colour Na K Cl Glu Prot
GeneXprt CLAT India ink Other
Sputum:
AFB GeneXprt PCP PCR
MC+S Other
Other (Fluid / BMAT / Biopsy):

Radiology and other (where indicated)
ECG:

X-Ray:

Ultrasound / Echo

CT Scan/ Other:

Admission / Disposition information
Patient disposition: D/C home Ward no: Step down (where):
Ward admission: Date: ____/____/_____ Time: ____h____ Not Applicable
Ward discharge: Date: ____/____/_____ Time: ____h____ Not Applicable
Death: Date: ____/____/_____ Time: ____h____ Not Applicable
ICU / HC Admission: Date: ____/____/_____ Time: ____h____ Not Applicable
ICU / HC Discharge: Date: ____/____/_____ Time: ____h____ Not Applicable
Final Assessment

Appendix B: Request for permission from hospital clinical director and H.O.D to conduct research study

To:

I am a consultant employed at the Charlotte Maxeke Johannesburg Academic Hospital Emergency Department and Critical Care Units. I hereby request permission to carry out a research study, which is a part requirement of a PhD project for which I am currently registered at the University of the Witwatersrand. My proposed research study is entitled: A Study of HIV Presentations to an Adult Emergency Department and Development and Validation of an Outcome Prediction Model (see attached protocol proposal)

If granted permission and after attaining ethical clearance by the Human Research Ethics Committee (Medical) of the University of the Witwatersrand, the study will take place at CMJAH. Informed consent will be obtained before enrolment in the study. At all times patient anonymity and confidentiality will be respected. The study will entail a prospective collection of data of patients presenting to CMJAH Emergency Department with HIV. The study will not incur any risk or expense to the patient or hospital. The study will need to include 800 patients and will run from 01/09/2016 to 31/03/2018. There will be no major expense involved. All minor expenses for the study will be borne by me. While undertaking this study I will ensure that my work and patient responsibilities will not be compromised. The exact details to patient / volunteer selection and approach can be found under the data collection section (p. 9 & 10) of my protocol proposal (see attached).

Yours Sincerely

___________________ 25/04/2016
Dr A E (Researcher) Date

Pending University of the Witwatersrand ethics review board approval, I hereby grant permission to Dr A Laher to carry out the research study outlined above.

_________________________ 25/04/2016
H.O.D Emergency Department Date

Appendix C: Patient information sheet

Good day, my name is. I am an emergency medicine physician and ICU specialist conducting a research project for my Doctor of Philosophy Degree (PhD). Thank you for taking the time to read this information sheet.

Study Title: A Study of HIV Presentations to an Adult Emergency Department and Development and Validation of an Outcome Prediction Model

Invitation to participate: I am inviting you to voluntarily take part in this research project.

Introduction: You presented to the hospital and are receiving treatment as per standard treatment guidelines. This is a study involving research and not routine care. Research is just the way we learn the answer to a question. My research project relates to HIV. In this study we want to learn more about why people with HIV present to an Emergency Department. We will also be studying adherence and compliance to HIV treatment and will also be developing a scoring system that will assist with determining outcomes in patients with HIV presenting to an Emergency Department.

What is involved in the study: Information for the study will be collected from your hospital records. If I require more information that is not found in your file, I may also ask you a few questions as well as examine you. Information that will be collected relate to your demographic details, medical findings, HIV related history and treatment compliance (where applicable), laboratory / other test findings and discharge / final outcome details.

Benefits: The information obtained may be useful for future patients as we may then be able to better understand HIV. This may also result in improved care for patients living with HIV.

Risks: This study would take approximately 15 minutes of your time. It is completely safe; there is no pain, no cost on your part and no side effects or risk of any harm.

Confidentiality: All your information is anonymous and cannot be later linked to you. Every effort will be made to keep your information confidential.

Participation is voluntary: You may choose to withdraw from the study at any time. Your treatment will not be affected if you take part, do not take part or at any time decide you no longer want to take part in the study.

Reimbursements: There will be no money paid to you for participating in the study. The results of the study will be available to you once the study has been completed.

Ethics: This study has been approved by the Human Research Ethics Committee (Medical) of the University of the WitwatersrandI really appreciate your time and help if you choose to volunteer for my research project.

Appendix D: Deferred patient information sheet (unable to get next of kin / legal guardian consent)

Study Title: A Study of HIV Presentations to an Adult Emergency Department and Development and Validation of an Outcome Prediction Model

Invitation to participate: I am inviting you to voluntarily take part in this research project.

Introduction: You were brought to the Charlotte Maxeke Johannesburg Academic Hospital Emergency Department on _____________. At that time you were not in a condition to volunteer to take part in this study. Now that your condition has been getting better, I would like ask you to volunteer your information regarding your illness for this study. This is a study involving research and not routine care. Research is just the way we learn the answer to a question. My research project relates to HIV. In this study we want to learn more about why people with HIV present to an Emergency Department. We will also be studying adherence and compliance to HIV treatment and will also be developing a scoring system that will assist with determining outcomes in patients with HIV presenting to an Emergency Department.

What is involved in the study: Information for the study was collected from your hospital records with the permission of the Human Research Ethics Committee (medical) of the University of the Witwatersrand. Information that was collected related to your demographic details, medical findings, HIV related history and treatment compliance (where applicable), laboratory / other test findings and discharge / final outcome details.

Benefits: The information obtained may be useful for future patients as we may then be able to better understand HIV. This may also result in improved care for patients living with HIV.

Risks: This study is completely safe; no cost on your part and no side effects or risk of any harm.

Confidentiality: All your information is anonymous and cannot be later linked to you. Every effort will be made to keep your information confidential.

Participation is voluntary: You may choose to withdraw from the study at any time. Your treatment was not and will not be affected if you take part, do not take part or at any time decide you no longer want to take part in the study. If you decide not to take part, your information will not be used for this study.

Reimbursements: There will be no money paid to you for participating in the study. The results of the study will be available to you once the study has been completed.

Ethics: This study has been approved by the Human Research Ethics Committee (Medical) of the University of the Witwatersrand.
I really appreciate your time and help if you choose to volunteer for my research project.

Appendix E: Deferred patient information sheet (with next of kin / legal guardian consent)

Study Title: A Study of HIV Presentations to an Adult Emergency Department and Development and Validation of an Outcome Prediction Model

Invitation to participate: I am inviting you to voluntarily take part in this research project.

What is involved in the study: Information for the study was collected from your hospital records with the permission of the Human Research Ethics Committee (medical) of the University of the Witwatersrand as well as from __________________ (name), who is your husband / wife / father / mother / son / daughter / other (specify): _______________. (See attached information sheet and signed consent form attached). Information regarding your HIV status was at no time told to him / her or anyone else, nor will it be told to anyone without your permission. Information that was collected related to your demographic details, medical findings, HIV related history and treatment compliance (where applicable), laboratory / other test findings and discharge / final outcomes details.

Benefits: The information obtained may be useful for future patients as we may then be able to better understand HIV. This may also result in improved care for patients living with HIV.

Risks: This study is completely safe; no cost on your part and no side effects or risk of any harm.

Confidentiality: All your information is anonymous and cannot be later linked to you. Every effort will be made to keep your information confidential.

Reimbursements: There will be no money paid to you for participating in the study. The results of the study will be available to you once the study has been completed.

Appendix F: Next of kin / legal guardian information sheet

I am an emergency medicine physician and ICU specialist conducting a research project for my Doctor of Philosophy Degree (PhD). Thank you for taking the time to read this information sheet.

Invitation to participate: Since you are the next of kin / relative of __________________________, I am inviting you to voluntarily take part in this research project on behalf of him / her.

Introduction: ___________________ (name of patient) presented to the hospital and is receiving treatment as per standard treatment guidelines. This is a study involving research and not routine care. Research is just the way we learn the answer to a question. In this study we want to learn more about the illness of your relative.

What is involved in the study: Information for the study will be collected from the hospital records of _______________________ (name of patient), who is your husband / wife / father / mother / son / daughter / other (specify): _______________. I may also ask you a few questions as well as examine ______________ (name of patient). Information that will be collected relates to demographic details, medical findings, laboratory / other test findings and discharge / final outcome details.

Benefits: The information obtained may be useful for future patients as we may then be able to better understand his / her illness. This may also result in improved care for other patients living with his / her illness.

Risks: This study would take approximately 15 minutes of your time. It is completely safe; there is no pain, no cost and no side effects or risk of any harm to you or the patient.

Confidentiality: All information is anonymous and cannot be later linked to you or the patients. Every effort will be made to keep all information confidential.

Participation is voluntary: You may choose to withdraw him / her from the study at any time. His / her treatment will not be affected by taking part, not take part or at any time deciding to no longer want to take part in the study.

Reimbursements: There will be no money paid to you or the patients for participating in the study. The results of the study will be available to the patient once the study has been completed.

Appendix G: Doctor / medical staff information sheet

My research topic is entitled: A Study of HIV Presentations to an Adult Emergency Department and Development and Validation of an Outcome Prediction Model

The aims of this PhD study are: 1) to describe the clinical presentation, pathology and outcomes of adult HIV positive patients presenting to the CMJAH ED, 2) to determining the underlying reasons for health deterioration of adult HIV positive patients presenting to the CMJAH ED (e.g. non-compliance with ART’s, unavailability of ART’s, resistance, education issues, etc.), 3) to develop an outcome prediction model for adult HIV positive patients presenting to the CMJAH ED, 4) to internally validate the outcome prediction model that will be developed. I will hope to achieve this via four concurrent studies, each concentrating on one of the aims.

Study design, population and site: This PhD will comprise 4 studies. Each of the aims listed above will comprise a study. All of these studies are prospective and will be conducted here at the Charlotte Maxeke Johannesburg Academic Hospital Emergency Department. I will require 800 patients for the entire study. It is estimated that data collection will continue over approximately one year.

What is required from you and your colleagues: Kindly inform me (contact details below) regarding any patients that present to this ED with confirmed or suspected HIV disease. Your assessment, management and documentation of findings should be comprehensive and goal directed as usual practice. You will not be required to complete any extra documentation or obtain study consent. Data collection and study consent procedures will be conducted by me from the hospital file and the patient. Information that will be collected relate to demographic details, medical findings, HIV related history and treatment compliance (where applicable), laboratory / other test findings and discharge / final outcome details.

Benefits: The information obtained may be useful for future patients as we may then be able to better understand and prognosticate patients presenting with HIV. This study may also result in improved care for other patients living with HIV. There will be no direct benefit to you or the patient.

Risks: This study is completely safe; there is no pain, no cost and no side effects or risk of any harm to you or the patient.

Confidentiality: All information is anonymous and cannot be later linked to you or the patients. Every effort will be made to keep all information confidential.

Participation is voluntary: If you choose not to get involved in the study, kindly inform me so that I can make other plans to identify potential study subjects. Your job and our working relationship will not be affected in any way if you choose to get involved, not to get involved, or at any time decide to withdraw involvement in the study.

Reimbursements: There will be no money paid to you or the patients for participating in the study.

Ethics: This study has been approved by the Human Research Ethics Committee (Medical) of the University of the Witwatersrand.

Sincerely yours

Appendix H: Consent form (patient)

Consent to act as a participant in research

I, __________________________________________________________, who is 18 years or older, consent / do not consent to participate in the research project entitled:

A Study of HIV Presentations to an Adult Emergency Department and Development and Validation of an Outcome Prediction Model

The procedures have been explained to me and I understand and appreciate their purpose, and the extent of my involvement.

I have read and understand the attached patient information leaflet.

I understand that the research forms part of a research project, and may not provide any direct benefit to me.

This study has been approved by the Human Research Ethics Committee (Medical) of the University of the Witwatersrand. Any queries may be directed to the chair; Professor P Cleaton-Jones on 011 717-1234

I am aware that my participation is voluntary, and that I am free to withdraw from the project at any time without any consequences.

_______________ ______________
Date Date

_______________ __________________
Participant Name Researcher

_______________ __________________
Participant Signature Researcher Signature

Researcher contact details:

Appendix I: Consent form (patient – deferred)

Consent to act as a participant in research

I, __________________________________________________________, who is 18 years or older, consent / do not consent for the information collected from my records thus to be used and I also consent / do not consent to continue to participate in this research project entitled:

A Study of HIV Presentations to an Adult Emergency Department and Development and Validation of an Outcome Prediction Model

The procedures have been explained to me and I understand and appreciate their purpose, and the extent of my involvement.

I have read and understand the attached patient information leaflet.

I understand that the research forms part of a research project, and may not provide any direct benefit to me.

I am aware that my participation is voluntary, and that I am free to withdraw from the project at any time without any consequences.

_______________ ______________
Date Date

_______________ __________________
Participant Name Researcher

_______________ __________________
Participant Signature Researcher Signature

Researcher contact details:

Appendix J: Consent form (next of kin / legal guardian)

Consent from next of kin / legal representative for participant in research

I, __________________________________________________________, who is 18 years or older, consent / do not consent to participate in the research project on behalf of: _________________________________________________

The procedures have been explained to me and I understand and appreciate their purpose, and the extent of my involvement.

I have read and understand the attached study information leaflet.

I understand that the research forms part of a research project, and may not provide any direct benefit to me or the patient.

I am aware that participation is voluntary, and that I am free to withdraw from the project at any time without any consequences.

_____________ ______________
Date Date

_________________________________ __________________
Next of kin / patient representative Name Researcher

____________________________________ __________________
Next of kin / patient representative Signature Researcher Signature

Researcher contact details:

Appendix K: Ethics clearance certificate

A Study of HIV Presentations to an Adult Emergency Department and Development and Validation of an Outcome Prediction Model

UBERESSAYS.ORG