The Impact of C-reactive Protein Testing

The Impact of C-reactive Protein Testing on Antibiotic Prescription in Febrile Patients Attending Primary Care in Low-resource Settings

PRIMARY OBJECTIVE To assess the impact of C-reactive protein (CRP) Point-of-care (POC) testing on health care worker prescribing behaviour in patients presenting to primary healthcare centres with an acute fever or recent history of fever.

SECONDARY OBJECTIVES To assess the impact of CRP testing on clinical outcomes within the 14 days of follow-up.

To assess the correlation between CRP results and clinical outcomes on the day 5 of the enrolment.

To estimate the impact of CRP testing on antibiotic consumption after first consultation.

To explore the attitudes of health centre staff towards the POC CRP test. To identify the prevalence of key pathogens in febrile patients in these settings.

To validate the ability of CRP to discriminate between viral and bacterial pathogens in a subset of patients with a microbiologically confirmed diagnosis.

Study Overview

• Status: Completed
• Conditions: Fever
• Intervention / Treatment: Other: No CRP will be measured onsite; Other: CRP cut-off of 20mg/L.; Other: CRP cut-off of 40mg/L.

Detailed Description

This is a multi-centre, individually randomized-controlled, three-armed pragmatic trial comparing CRP guided antibiotic prescription in febrile patients to the current standard prescribing systems. The study will be implemented in low and middle-income countries in tropical settings, including Myanmar and Thailand in the first instance.

As participation in the study itself may raise awareness amongst the health centre staff and local community, current antibiotic prescribing rates will be assessed in an observational run-in period before the intervention. Health workers will be asked to complete case record forms (CRF) for patients presenting to their health centre with current or recent fever. This aims to capture current antibiotic prescribing habits before the study is launched.

During the intervention phase, patients fulfilling the inclusion criteria and who consent to participate in the study will be randomised to one of the three arms, as follows:

1. Control group: The health care provider will manage the patient using standard guidelines. No CRP will be measured onsite
2. Group A: CRP will be measured by a study nurse and the result will be communicated to the health care provider as "High-CRP" or "Low-CRP" using a low CRP cut-off of 20mg/L.
3. Group B: CRP will be measured by a study nurse and the result will be communicated to the health care provider as "High-CRP" or "Low-CRP" using the higher CRP cut-off, of 40mg/L.

For the two intervention arms, the following guidance will be given to the health care provider: if the CRP test is reported as 'high', antibiotic treatment is recommended, following local guidelines, and if it is reported as 'low' antibiotics are not recommended. In either case the information provided by the CRP test should be interpreted alongside their clinical judgement.

A second CRP will be sampled at day 5 of the follow-up (+/- 1 day) for all the patients, using capillary blood from a finger prick.

The investigators will use the NycoCard Reader II (Axis Shield, Norway or equivalent) to measure CRP levels.

A venous blood sample and a nasopharyngeal swab will be taken in the control group at enrolment and sent to a central laboratory in order to detect the presence of the following key pathogens by real-time polymerase chain reaction (PCR):

• Flavivirus
• Alphavirus
• Influenza A & B
• Rickettsia including typhus group and spotted fever group
• Leptospirosis
• PCR 16s for the detection of any bacteria.
• Malaria

CRP will also be retrospectively measured in these blood samples to validate its ability to distinguish between viral and bacterial infections.

A urine sample will be collected to detect the presence of antimicrobials at day 0 and day 5. This procedure will ascertain pre-study antibiotic intake, as well as the patient's compliance to the health care worker's prescription or advice that antibiotics are not required, during their participation in the study. The urine samples will be collected at the sites, divided into aliquots and frozen to -80°c to be stored on site. Monthly shipments will be made to the laboratory at Mahidol Oxford Tropical Research Medicine Unit in Thailand for analysis.

The investigators aim to follow-up every patient face to face on day 5 (+/-1 day), and either by phone or face-to-face interviews 14 days (+/-2 days) after the initial visit.

• Study Type: Interventional
• Enrollment (Actual): 2410
• Phase: Not Applicable

Contacts and Locations

Study Locations

• Thailand
  • Chiangrai, Thailand, 50007
    • Chiangrai Clinical research Unit

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: 1 year and older (Child, Adult, Older Adult)
• Accepts Healthy Volunteers: No
• Genders Eligible for Study: All

Description

Inclusion Criteria:

• Patients aged ≥1 year presenting to selected primary healthcare centres
• Tympanic temperature >37.5°c or history of fever ≤ 14 days.

Exclusion Criteria:

• Patients that in view of the study nurse are in need of emergency referral to a higher-level facility, as indicated by either 1) impaired consciousness or 2) inability to take oral medication.
• In sites that routinely test for malaria, patients with a positive malaria rapid diagnostic test or microscopy will be excluded
• The main complaint is a trauma and/or injury
• Suspicion of tuberculosis (any medical history and/or physical examination suggesting tuberculosis)
• Suspicion of Urinary Tract Infections (any medical history and/or physical examination suggesting urinary tract infections)
• Suspicion of local skin/dental abscess (any medical history and/or physical examination suggesting a local skin/dental abscess
• Any presenting symptom present for more than 14 days
• Bleeding, including otorrhagia, haematemesis, haemoptysis, haemorrhagic petechiae, haematuria, bloody diarrhoea.
• Not able to comply with the follow-up at Day 5 (+ / - 1 day).

Study Plan

How is the study designed?

Design Details

• Primary Purpose: Other
• Allocation: Randomized
• Interventional Model: Parallel Assignment
• Masking: Double

Number of Arms

3

Arms and Interventions

Participant Group / Arm	Intervention / Treatment
Other: CRP-Control; The health care provider will manage the patient using standard guidelines. No CRP will be measured onsite	Other: No CRP will be measured onsite; No CRP will be measured onsite
Other: CRP-A; CRP will be measured by a study nurse onsite and the result will be communicated to the health care provider.	Other: CRP cut-off of 20mg/L.; Health care worker will be given an advice to prescribe antibiotic to patient who has CRP lever < 20mg/L.
Other: CRP-B; CRP will be measured by a study nurse onsite and the result will be communicated to the health care provider.	Other: CRP cut-off of 40mg/L.; Health care worker will be given an advice to prescribe antibiotic to patient who has CRP lever < 40mg/L.

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
The proportion of recruited patients prescribed an antibiotic	The proportion of recruited patients prescribed an antibiotic at the health centre on or between their enrolment and the first follow-up visit (on day 5 +/- 1 day).	6 Days

Secondary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Duration of symptoms	Symptoms as defined by fever, any other symptom	14 Days
Severity of symptoms	Severity will be assessed by the nurse using a rating scale graded from 1 to 4 (#Grading 1=mild, 2=moderate, 3=severe, 4=potentially life-threatening), according to the patients' description of their symptoms, and the findings of the physical examination.	14 Days
Frequency of severe clinical outcomes	as defined by admission to hospital or death within the 14 days of follow up	14 Days
Proportion of patients that needed their clinical management changed within the 14 days of follow-up.		14 Days
Frequency of unplanned re-consultation		14 Days
Clinical outcome on Day 5 compare to CRP level		5 Days
Proportion of patients with an immediate versus subsequent prescription within 2 weeks.		14 Days
Proportion of urine samples that tested positive for presence of antibiotics		14 Days
Interview of health centre staff for attitudes and satisfaction	Attitudes and satisfaction of health centre staff towards the CRP POC test	6 months
The proportion of a population who have key pathogens	Key pathogens are Flavivirus, Alphavirus, Influenza A & B, Rickettsia including typhus group and spotted fever group, Leptospirosis, PCR 16s for the detection of any bacteria and malaria	6 months
Percentage of subjects who are correctly identified as having bacterial infections by using CRP testing	Sensitivity is defined as the proportion of patients with high CRP, out of all PCR confirmed bacterial infections.	6 months
Percentage of subjects who are correctly identified as having viral infections by using CRP testing	Specificity is defined as the proportion of patients with low CRP, out of all PCR confirmed viral infections.	6 months
The likelihood ratio (LR) of correctly identifying a bacterial infection by using CRP-testing.	The positive likelihood ratio is the fraction of sensitivity over (1 - Specificity) , and the negative likelihood ratio is the fraction of (1 - Sensitivity) over Specificity.	6 months
The receiver operating characteristic (ROC) for evaluating the accuracy of CRP-testing in identifying bacterial infections.	We will use the ROC, or ROC curve, as a graphical plot, in order to illustrate the performance of the CRP-testing to classify bacterial and non-bacterial infections as its discrimination threshold is varied.	6 months

Collaborators and Investigators

• Sponsor: University of Oxford
• Collaborators: Department of Medical Research, Lower Myanmar
• Investigators: Principal Investigator: Yoel Lubell, MD, Mahidol Oxford Tropical Medicine Research Unit

Publications and helpful links

General Publications

• Haenssgen MJ, Charoenboon N, Do NTT, Althaus T, Khine Zaw Y, Wertheim HFL, Lubell Y. How context can impact clinical trials: a multi-country qualitative case study comparison of diagnostic biomarker test interventions. Trials. 2019 Feb 8;20(1):111. doi: 10.1186/s13063-019-3215-9.
• Althaus T, Thaipadungpanit J, Greer RC, Swe MMM, Dittrich S, Peerawaranun P, Smit PW, Wangrangsimakul T, Blacksell S, Winchell JM, Diaz MH, Day NPJ, Smithuis F, Turner P, Lubell Y. Causes of fever in primary care in Southeast Asia and the performance of C-reactive protein in discriminating bacterial from viral pathogens. Int J Infect Dis. 2020 Jul;96:334-342. doi: 10.1016/j.ijid.2020.05.016. Epub 2020 May 11.
• Althaus T, Greer RC, Swe MMM, Cohen J, Tun NN, Heaton J, Nedsuwan S, Intralawan D, Sumpradit N, Dittrich S, Doran Z, Waithira N, Thu HM, Win H, Thaipadungpanit J, Srilohasin P, Mukaka M, Smit PW, Charoenboon EN, Haenssgen MJ, Wangrangsimakul T, Blacksell S, Limmathurotsakul D, Day N, Smithuis F, Lubell Y. Effect of point-of-care C-reactive protein testing on antibiotic prescription in febrile patients attending primary care in Thailand and Myanmar: an open-label, randomised, controlled trial. Lancet Glob Health. 2019 Jan;7(1):e119-e131. doi: 10.1016/S2214-109X(18)30444-3.

Study record dates

Study Major Dates

• Study Start (Actual): June 8, 2016
• Primary Completion (Actual): September 30, 2017
• Study Completion (Actual): September 30, 2017

Study Registration Dates

• First Submitted: March 10, 2016
• First Submitted That Met QC Criteria: April 27, 2016
• First Posted (Estimate): May 3, 2016

Study Record Updates

• Last Update Posted (Actual): August 15, 2018
• Last Update Submitted That Met QC Criteria: August 14, 2018
• Last Verified: August 1, 2018

More Information

Terms related to this study

• Keywords: antibiotic; C-reactive protein
• Other Study ID Numbers: CRP POC

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: NO