[Proposed Experiment]: Health Malawi - Research Protocol Stage 1

Sunny Wu
Submitted by
Sunny Wu | 2 weeks ago | in Health Consulting

Lean Phase: Solution

Assumption:  385 patients can correctly and accurately identify their symptoms on the paper questionnaire in comparison with what they verbally identify to the clinician.

Time Frame: 2 -  3 weeks 

Success Metric: 80% of data points accurately identify symptoms on both paper form and verbally.

Green light: Go on to Research Protocol 2

Orange light: If 70% achieve accuracy of detailing symptoms on both paper form and verbally then reassess and revision of questionnaire is required.

Red light: If less than 70% achieve accuracy of detailing symptoms on both paper form and verbally then reassess and revision of questionnaire and symptom identification poster is required.

Experiment Build

As part of Research Protocol 1, the research team intends to visit Zingwangwa Clinic and Gateway Clinic to obtain 385 data points using a paper based survey. The collection of this data constitutes ‘Stage One’ of the research protocol. This paper based symptom data collection method is critical as it allows the close control of data collection methodologies laid out in the Data and Recruitment SOP’s.

Before we can begin the use of any USSD symptom checker/triage system, it is important that we first understand the level of medical literacy that patients have. To ensure this process is controlled, paper surveys are to be used first.

A paper survey is to be filled out by both a patient and a clinician, at separate times. The survey will involve information such as age, sex, symptoms and the purpose of their visit. Through comparing the results of the patient survey to the clinician survey, we will be able to identify how well people can identify their own symptoms. This understanding is important as it provides insight into how successfully people will be able to input their symptoms on any potential USSD systems. A digital algorithm/ system has been developed so that all of the data collected from the 385 can be analysed and specific correlation between what the patients communicate to the research assistants and what they communicate to their clinicians can be determined. The paper surveys will be useful in giving us a solid scientific backing to get the data points to feed into our eventual symptom checker and test its accuracy.

385 data points has been deducted as the minimum sample size of the population of people that ‘attend health centres’ and will allow for stratification of data into age groups. This range of ages means random sampling will be difficult to conduct, but it will mean that the overall pool of data will reflect the population. Once this target is reached we can begin to design the first stages of a digital triage system. After collecting 385 data points, the research team will then progress to ‘Stage Two’.



edited on 8th February 2019, 07:02 by Sunny Wu
No comments yet, be the first to post one!