Prospective daily diary study reporting of any and all symptoms in healthy adults in Pakistan: prevalence and response

Abstract

ObjectivesPrevalence of symptoms in everyday life and how people respond to these symptoms is little studied outside Western culture and developed countries. We sought to use modified diary methods to explore the prevalence of and responses to symptoms in Pakistan.DesignProspective daily survey of symptoms and response.Setting8 cities across four provinces in Pakistan.ParticipantsStratified intercept in each city to recruit 153 participants of which 151 completed.Primary and secondary outcome measuresEach day for 30 days, participants were prompted by text message (short message service (SMS)) to complete a symptom diary. On days where symptoms were experienced, participants also reported how they responded. Prevalence was adjusted to population age and gender distributions.Results92% of participants experienced symptoms (adjusted prevalence 94%, 95% CI 91% to 97%), with musculoskeletal pain (83%, adj. 84%, 95% CI 84% to 90%) and respiratory symptoms (75%, adj. 77%, 95% CI 71% to 84%) the most prevalent types of symptoms. Self-medication and use of home remedies and traditional medicines were the most common responses. Seeking professional help or using conventional medicine were less common, and self-medication responses included the use of antibiotics without prescription. The range of home remedies and traditional medicines was very diverse.ConclusionsWhile symptom experience in Pakistan was similar to Western countries, home remedies were much more frequently used to respond to symptoms. Understanding how people respond and manage their experience of symptoms outside formal healthcare is important for designing effective policy and interventions, and this needs to be understood within the broader context including the cultural and economic setting, the health system and other structural determinants of health.

Too many zeros and/or highly skewed? A tutorial on modelling health behaviour as count data with Poisson and negative binomial regression

“…The data from Staunton et al ( 2015 ) did not involve human participants but was based on publicly available data. For Anwar et al ( 2017 ) ethical approval was received from the University of the Punjab (HEC/UCP/1916A) in Pakistan, and we also received approval from the University of Otago Human Ethics Committee (F12/008).…”

“…This data comes from Anwar et al ( 2017 ) with the outcome variable (number of visits to a health professional in 30 days) originally dichotomised into visited or not visited, and modelled with logistic regression. Instead here, we will treat it as the count variable it is.…”

“…Left panel is the number of views on YouTube of videos about scoliosis (Staunton et al, 2015 ). Right panel is the number of days on which participants sought help from a health professional over 30 days (Anwar et al, 2017 ).…”

“…Similarly in a related study, Veldhuizen, Atsma, van Dongen, and Kort ( 2012 ) looked at stopping blood donation (Yes/No). I have also looked at whether people in Pakistan visited a health professional over a month (visited/did no visit), when it would have been possible to analyse the number of visits across the month (histogram of visits displayed in the second panel of Figure 1 ; Anwar, Green, Norris, & Bukhari, 2017 ). In these three studies, logistic regression was used.…”

Too many zeros and/or highly skewed? A tutorial on modelling health behaviour as count data with Poisson and negative binomial regression

“…Similarly in a related study, Veldhuizen, Atsma, van Dongen, and Kort (2012) looked at stopping blood donation (Yes/No). I have also looked at whether people in Pakistan visited a health professional over a month (visited/did no visit), when it would have been possible to analyse the number of visits across the month (histogram of visits displayed in the second panel of Figure 1; Anwar, Green, Norris, & Bukhari, 2017). In these three studies, logistic regression was used.…”