Sunset at MRRH 8/1/2018
bolded terms have additional information found at the bottom of the section.
Mbarara Regional Referral Hospital (MRRH) is about 250 kilometers southwest of the capital city Kampla.
The hospital severes the surrounding ten districts in Southwestern Uganda.
The hospital frequently treats patients from the neighboring countries like Rwanda and Tanzania.
Hospital bed capacity is 350.
There are four operating theatres, neurosurgery has access to one 3-4 times per week.
There is a radiology department. They perform CTs. A CT costs $45 USD. Patients needing MRI are sent to kampala (5 hour drive).
There is an intensive care unit (ICU). They have one ventilator so ICU capcity is one.
There are 4 general surgeons, 2 pediatric surgeons, a cardiothoracic surgeon, 2 plastic surgeons, one orthopaedic surgeon, one neurosurgeon, and 19 general surgery residents (senior housing officers). See key for details on provider titles.
There are 6 anaesthesiologists, 3 anaesthetic officers and 7 anaesthesia residents.
There are an additional 2 medical officers (general practitioners) and ~15 intern doctors.
$45 USD - This cost is very high for patients. Some families will sell land to afford CT. The hospital has a voucher system to provide free CTs but to qualify is exceedingly difficult.
Key - The medical education system is set up as follows. At 18, Ugandans go to medical school for 5 years. At completetion, they are referred to as a doctor. They proceed to a one year internship rotating through all hospital departments ( interns ). Next they work at a hospital as a medical officer. This can last from a couple years to life. A medical officer does everything: basic surgeries, pediatrics, IM, FM, OB/GYN. If desired, a medical officer returns to residency to specialize. They are called a senior housing officer at this time. Upon completion they are referred to as a consultant.
They lack the needed equipment to perform surgeries when needed.
Coordinating surgeries with anaesthesiology is a major barrier.
Sending a patient to the ICU is not a purely clinical decision.
There is one neurosurgeon (Dr. Kitya) and one rotating general surgery resident changing every 6 weeks.
There is limited OR / theatre space for neurosurgery
equipment - The neurosurgery team lacks certain needed drills and accessories to perform the procedures which are needed. For example, the pedal for the drill broke which postponed surgeries for multiple days. MRRH had to borrow and ship a pedal from another hospital 6 hours away.
anesthesiology - This challenge may be the biggest barrier to timely surgical care. Securing an anesthesiologist for a surgery is a political process. Anesthesiologist can refuse to do a surgery, can tell the neurosurgery team they will do two of the three requested cases or tell neurosurgery that this patient does not need a surgery. Neurosurgery must lobby with anesthesiologists each day why a patient needs surgery.
ICU - There is only one equipped ICU room. There are 7 other ICU room but none of these 7 has a ventilator. Similar to surgery, neurosurgery must convince the ICU team to accept a patient. The ICU team can refuse patients. For example, neurosurgery lobbied to have a patient with a GCS of 4 sent to the ICU. ICU eventually accepted but told the team you are simply trying to ruin our mortality rate. In another example, a patient with a GCS of 5 which needed ICU care was not accepted to the ICU and died a few days later in the emergency ward.
theatre space - Neurosurgery has access to one of four theatre room about 3-4 times per week. The neurosurgery team will transport patients to other nearby hospitals with theatre space about once a week.
The borrowed pedal for the drill
There is no electronic medical record ( EMR) for the hospital.
There are no computers used at the hospital.
There is no freely available wifi at the hospital.
The power goes out at the hospital 2-3 times per week.
EMR - The neurosurgery department has an on-going registry which started summer 2016. Detailed information on this registry can be found below.
computers - No department or ward in the hospital has computers. MRRH uses paper paper charts. A dedicated research assistant for the neurosurgery department uses a laptop to maintain the neurosurgery registry.
wifi - Some people will bring a personal modum or mifi device (a really small modum) to an office for work. The cost is about $30 USD for 10 Gb for one month. The connection is above average. When the power goes out, the wifi 95% of the time will continue to work.
power - The hospital will use back up generators when the power is out. Surgeries are not performed when the power is out.
Portable wifi device. Very effective.
There is a neurosurgery registry at MRRH which started summer 2016.
There is one research assistant who enters data into registry.
The data quality is a challenge.
neurosurgery registry - This includes all neurosurgical conditions, adult and pediatric patients, inpatients and outpatients. The dataset has about 1,000 patients with 225 variables.
research assistant (RA) - The RA flips through the patient chart entering the patient information into a microsoft access database. There is no form used to collect the patient info in the chart. As result, the RA extracts the information page by page from the notes of the chart. This takes 15-45min per chart (in rare cases a chart for a long hospital stay requires one day to upload).
data quality - There are 50% or greater missing data for vitals and for GOS. Vitals - there are no bp cuffs, pulse oximiters, thermometers available in the hospital. The doctors must purchase their own. If doc does not have equipment, they will not record this info in the chart. The GOS is not routinely performed at discharge. If the RA is present, she must remind doctors to do determine the GOS. 90%+ of the recorded GOS are a 5, and there are no 2s or 3s. They do not calculate GOSe.
The feedback on GRYMTR was positive. Detailed feedback is found in a seperate document.
There is demand for maintaining a running list of neurosurgery / TBI patients in the ED.
Expressed potential uses of GRYMTR were lobbying for patient care, setting priorities, facilitating communication with families.
The doctors may not be the target end-user of the tool but rather a consumer of the output.
Mindset, ethos of the hospital, is possibly accepting of a decision support tool.
running list - This list would present all neurosurgery patients currently in the ED. Occasionally a neurosurgery patient will be in the ED without the team’s knowledge. Tracking down the patient chart is cumbersome and at times not possible. A central list, with organized vitals, tracking interventions provided, and displaying the risk score, would be welcome.
potential uses - As mentioned eariler, securing ICU and theatre space is a highly political process. An argument to anesthesiology or the ICU team based on a risk score with or without the intervention could be superior to GCS alone. There are at times 5+ patients needing surgery, the tool could help in prioritizing cases. However, there is rarely 5 patients presenting to the ED at the same time needing surgery. Over the course of a week, patient will trickly in needing cases, thus GRYMTR would have to be repurposed to generate risk scores for patients already in the hosptial. Finally, a risk score could help convince families that surgery should not be pursued for a patient.
target end-user - Feedback - “If you want to be sure for this tool to fail, have the intern be the end-user.” This comment reflects a doctor’s concern that collecting vitals for the tool will take too much time. Entering information in an app for a couple minutes is a major demand on an already over-stressed schedule. Feedback - “This tool is easy, anyone could use it” - This comment by a doctor was to highlight a lower-skilled team member can be trained to enter the info and report the risk score to the team.
possibly accepting - The feedback was positive from resident physicians. There is a learning curve to overcome to convince Dr. Kitya, department head for neurosurgery, to allow use of this tool.
IRB approval and hospital approval are successfully obtained.
UNCST submission will occur mid-August.
The doctor interviews and DCE will occur November 2018 at MRRH.
During my return trip, I will spend 2-3 weeks at Mulago Hospital in Kampala to increase sample size.
I will begin adapting the risk calculator from Tanzania to MRRH.
Explore hiring RA to pilot GRYMTR in ED.
I will return to Uganda around Oct 17th through beginning of January. We have a great opportunity, with IRB approval and department support, to collect needed data on the clinical application of GRYMTR!!
doctor interviews - Josephine and a general surgery residnet have offered to help in recruiting docs and conducting interviews.
sample size - Josephine has offered to join me in Kampala to help in recruiting doctors.
adapting - the goal is to complete this step prior to my return trip mid october. This will be challenging given the inconsistent collection of vital sign readings and GOS scores at MRRH.
hiring RA - One resident recommended hiring an RA to collect the information GRYMTR and generate the risk score. To give the pilot the greatest chance for success, he recommended not to rely on the doctors to use the tool.
The neurosurgery team at MRRH. Agandi!! (hello)