Treatment-seeking behavior and delay in children with severe malaria in Uganda: results of a Markov analysis


Autoria(s): Walters, Kelly Marie
Contribuinte(s)

Flaxman, Abraham

Data(s)

14/07/2016

01/06/2016

Resumo

Thesis (Master's)--University of Washington, 2016-06

Background Malaria is a treatable disease yet is a leading cause of death of children in Uganda. Children who experience ACT treatment delay of 24 hours or more are in danger of the progression of uncomplicated malaria to severe, life-threatening malaria. Previous research has examined populations that are more at risk of experiencing delay, and types of sources accessed by caregivers of children with malaria, but no quantitative model has been previous used to examine the relationship between treatment-seeking actions taken, delay, and disease progression. Methods Caregivers of children (n = 325) with severe malaria presenting to Jinja Children’s Hospital were interviewed about their treatment-seeking itineraries from onset of fever up until hospital enrollment. The actions taken at each step and the exact time at which they occurred were used to create a continuous-time, multi-state Markov model. Frequency of each action, average time spent on each action, and probabilities of taking each action given the last action taken were calculated. Two hypothetical counterfactual scenarios, one in which visiting drug shops was not allowed (counterfactual 1), and another in which drug shop visits were considered perfectly effective (counterfactual 2), were simulated, and the proportion of delayed cases was calculated in each scenario. Results Of cases analyzed (314), 33.4% visited a drug shop at some point in their treatment-seeking itinerary. Almost half (48%) of all drug shop visitors went home to self-medicate as their next step. Staying home and staying home to self-medicate contributed the greatest amount of time (30 hours and 35 hours, respectively) of any action. Of the real itineraries analyzed, 86.0% of cases experienced treatment delay of greater than 24 hours. In counterfactual 1, this number dropped to 70.1% of cases experiencing delay, and in counterfactual 2, 65.3% of cases experienced delay. Conclusions Drug shops played a substantial role in the delay of appropriate treatment among cases with severe malaria eventually presenting to a regional referral hospital. Interventions that address both drug shop use and drug shop effectiveness have great potential to decrease delay and severe malaria progression on a population level.

Formato

application/pdf

Identificador

Walters_washington_0250O_16202.pdf

http://hdl.handle.net/1773/36426

Idioma(s)

en_US

Palavras-Chave #Behavior #Malaria #Markov model #Treatment #Uganda #Public health #global health
Tipo

Thesis