Determinants of discharge against medical advice from a rural neurosurgical service in a developing country: A prospective observational study.

BACKGROUND: In low-resource regions of the world, discharge against medical advice (DAMA) is one empiric contributory factor to poor in-hospital outcome that is not often mentioned. This study aims to investigate the determinants of DAMA from a rural neurosurgical service in a developing country.
METHODS: This was a prospective observational study of all patients who discharged against medical advice in our service between November 2018 and October 2019.
RESULTS: There were 88 patients, 67 (76.1%) males, in the study, (M:F = 3.2:1), representing 17.4% of our patient population in the study period. The peak incidence was in the 20-29 years age group which accounted for 37.5% of the cases. About 55% of the patients presented directly to our center; 31.8% were referred from other hospitals, while 3.4% came from traditional caregivers and 1.1% from religious homes. Head injury was the most common indication for presentation (76.1% of the cases). The duration of hospital stay ranged from 2 h to 14 days. Majority of the patients (87.5%) left the hospital within 8 h of presentation. The reason for DAMA was financial constraints in 50% of cases, inadequate health literacy in 20.5%, financial constraints and poor health literacy together in 12.5%, religious misgivings in 4.5%, and traditional belief in 2.3%. Neurotrauma was predictive of early DAMA (P = 0.001).
CONCLUSION: The rate of DAMA was high in our study. Financial constraints with other socioeconomic limitations were the most common causes of DAMA in our environment. Copyright:

INTRODUCTION

Discharge against medical advice (DAMA) is an issue generating some significant concerns in health systems globally.[11,13,14,29] It is a prominent albeit less often discussed cause of poor heath outcome from hospitalized patients. It may result in inadequate treatment and increased rate of hospital readmission, complications, and mortality.[12,17,26,28,49] The rate of DAMA is approximately 1–2% of all hospital discharges in most of the literature from the West but may range from 0.3 to 25.9% depending on the setting of the study, social history of the patients, and medical condition among other factors.[8,17,33,50,51] Note worthily, higher rates of DAMA in developing than developed countries have been documented by some authors.[9,12,35,39,44]

Several causes of DAMA have been reported, including financial difficulties/lack of health insurance; dissatisfaction with the care received, and/or, the hospital environment; lack of access to skilled and qualified physicians, lack of advanced medical services; and presumed unsatisfactory results from treatment, and presumed unduly prolonged in-hospital stay of patients.[1,8,20,23,33,35,42] The problem of DAMA is probably more significant in developing countries with fewer medical specialists,[2] prevailing poor economic power, and, particularly, a global lack of health insurance necessitating out-of-pocket payment for health-care expenses in most cases.[4] This is even more so for neurosurgical care which notably has a much higher in-hospital costs of care compared to other disease conditions.[4] In Nigeria, the cost of brain CT/MRI and daily intensive care unit – admission charges are at least double the national minimum wage per month, while the cost of neurosurgical operative procedures is more than the national minimum wage per annum in most cases.

The previous studies on DAMA in our country as well as most studies from other parts of the world were in mixed patients’ populations in the medical, psychiatric, or surgical/ orthopedic wards.[1,9,12,17,23,28,29,49,50] None such that we are aware of has been reported from neurosurgical service, not globally, and certainly not from sub-Saharan Africa. This study, therefore, aims to explore for the rate and determinants of DAMA in neurosurgical patients in a rural tertiary care hospital of a resource-constrained African developing country.

MATERIALS AND METHODS

This was a cross-sectional study of all prospective patients who discharged against medical advice in our neurosurgical service over a 12-month period from November 2018 to October 2019. Our neurosurgical practice is a newly inaugurated one in a tertiary level hospital in a rural part of Nigeria, a large country in sub-Saharan Africa. We collected data on the demographic characteristics of the patients, the diagnosis, duration of symptoms, duration of stay in our service before the DAMA, reasons for the DAMA, plus representation in the hospital after the DAMA, as well as the reasons for the representation and the interval between the DAMA and the repeat presentation. Early DAMA (DAMAe) was defined as DAMA within 8 h of admission and late DAMA (DAMAd) as DAMA after 8 h.[31] Data were analyzed with the Statistical Package for the Social Sciences (SPSS) version 20 (SPSS Science Inc., Chicago, IL, USA) and presented in descriptive categorical variables and proportions. The Pearson’s Chi-square test of associations was used to determine variables associated with early DAMA. Multivariate logistic regression analysis was modeled to determine independent predictors of early DAMA. Variables that attain statistical significance on bivariate analysis were inputted into the regression model. The level of significance for all statistical analyses was set at P < 0.05.

RESULTS

There were 88 patients in the study, 67 (76.1%) males and 21 females (M:F = 3.2:1), representing 17.4% of the 506 cases seen in our neurosurgical practice during the study period [Table 1]. The mean age was 33.4 (SD17.7) years and ranged from 35 h to 93 years. The peak incidence was in the 20–29 years age group which accounted for 37.5% of the cases [Figure 1]. Only 10.2% of the patients were children, aged 35 h–17 years. The duration of symptoms ranged from 30 min to 20 years. More than three-fifths (62.5%) of the patients presented directly to our center, 31.8% were referred from other hospitals; 3.4% came from traditional caregivers and 1.1% each from religious and nursing homes [Figure 2]. Head injury was the diagnosis in 76.1% of the cases, spinal cord injury in 9.1%, and myelomeningocele and chronic subdural hematoma each in 2.3%. Some 1.1% had both head injury and spinal cord injury while brain tumor, hemorrhagic stroke, cerebellar abscess, hydrocephalus, and degenerative spine disease each accounted for 1.1% of the cases [Table 2]. All patients funded their health care with us out-of-pocket: none had health insurance coverage.

Table 1:

Clinical and demographic characteristics of the patients.

Figure 1:

Age distribution of the patients.

Figure 2:

First point of care of the patients.

Table 2:

Summary of cases.

The median duration of hospital stay in this cohort of patients that discharged AMA was 6.5 h (range = 2 h–14 days). The DAMA was early, within 8 h of admission (DAMAe) in 87.5%, and late, after 8 h (DAMAd) in 12.5%. The reason for DAMA was low income in 50% of cases, low health literacy in 20.5%, low income and health literacy together in 12.5%, long distance in 8.0%, and conflicting religious and traditional beliefs in 4.5% and 2.3%, respectively [Table 3]. Following a bivariate analysis, age <50 years (P = 0.02), male gender (P = 0.01), neurotrauma (P < 0.01), and the presence of associated injuries (P = 0.02) attained statistical significance as predictors of early DAMA [Table 4]. The significant variables were selected for multivariate regression analysis. Based on this final model, the only significant risk factors for early DAMA in this study were neurotrauma (P < 0.01) [Table 5]. Only 4 patients (4.5%) represented to our hospitals after the DAMA, 2 (50%) because they were compelled by relatives to do so, and the other two because of neurological deterioration. The interval from the DAMA to representation ranged from 1 h to 18 days.

Table 3:

Summary of indications for discharge against medical advice.

Table 4:

Relationship between clinical and demographic status and DAMA.

Table 5:

Independent predictor of early discharge against medical advice.

DISCUSSION

DAMA is a well-recognized concern in in-patient health care globally with significant attendant negative economic and health consequences.[10,11,14,29] There is hardly any data-driven study known to us about DAMA in a purely neurosurgical patients’ population globally, let alone in developing countries in particular. This study interrogates the burden and the determinants of this problem in a cohort of neurosurgical patients’ population in a Nigerian rural practice setting. About 17.4% rate of DAMA in our study from this neurosurgical patient cohort was higher than 0.5–2.2% reported in other in-hospital patients from the developed world,[26,43,44,45,50] and 0.002–5.7% from previous reports from our country.[9,23,27,35,37,48] A relatively higher rate of 7.2% had also been reported in a population of medical in- patients by Fadare et al. working in an environment similar to ours from the same region of our country.[24] In a cohort of pure neurological patients in India, Lakhotia et al. reported a DAMA rate of 14.9%.[31]

The predominance of male gender in this study is in agreement with trends in other studies locally and abroad.[7,23,31,34,36,38,44] Female predominance, however, was reported by Jimoh et al.[29] and Hasan et al.[25] More than one- third (37.5%) of our patients were between 20 and 29 years. The predominance of patients in the economically productive age groups and the aforementioned male predominance in this study may be due to the predominance of neurotrauma patients in our study population since it is known that trauma is the leading cause of neurosurgical diseases in these groups.[3,5,6,22,30] The most common reason for DAMA in this study was financial constraints, occurring alone in 50% of the cases and in combination with low health literacy in another 12.5%. This finding has been widely reported by other authors both in the developing and developed countries.[13,24,25,29] Poor socioeconomic status and lack of health insurance are probable reasons for this.[26,28,29,40] Our figure of about 63% regarding the contribution of economic hardships for the cause of DAMA is higher than 32.9% reported by Jimoh et al.,[29] 40.6% by Hasan et al.,[25] and 48% by Fadare et al.,[24] but similar to 61.1% in the study by Lakhotia et al.[31] The variation between our figures and that of Fadare et al.[24] who were working in a setting similar to ours may be due to the significantly higher costs of in-hospital care of neurosurgical patients. Notably, none of the patients in this series had health insurance cover.

Other reasons for DAMA identified in this study were poor medical insight and the need for proximity to relatives/ caregivers in some of the patients. About 33% of the patients discharged against medical advice because they felt that they were well while 8% left because of their felt need for proximity to relatives/caregivers despite no access to neurosurgical service at these destinations. These factors have all been documented in the literature.[13,25,29]

Notably in this study, none of the patients discharged against medical advice because of dissatisfaction with the care received, or the medical staff, or the hospital environment. This finding is contrary to those of other reports which found a proportion of their patients leaving the hospital against medical advice because of dissatisfaction with the medical staff or the hospital environment.[8,15,16,18,19,32,47] This may be due to the small sample size of this study subjects or the fact that the prevailing poor economic status and poor level of education of the patients left them with little power of a choice and rather made them accept whatever was offered to them. Another probable reason is the paucity of neurosurgeons and hence neurosurgical services in our country and the concentration of the available ones in the urban areas, thereby limiting the options available to the poor rural dwellers.[41]

DAMA because of religious and traditional beliefs was a rather interesting finding in this study. Although the percentage was low in this study (6.8%) despite the prevalence of these two factors in our environment, this finding highlights the challenge these beliefs pose to health care in poor societies with high level of poor health literacy, particularly in the developing world.

More than four-fifth of our patients (87.5%) in this study were of the early DAMA status, leaving the hospital within 8 h of presentation. This figure is significantly higher than the 25.7% reported by Lakhotia et al.[31] Similar to their study, however, we found no significant association between the age of the patients and early DAMA. However, contrary to their report, the gender of the patient was not predictive of early DAMA in our own study, whereas a diagnosis of neurotrauma was significantly associated with early DAMA in this study. This may be due to the predominance of these cases in neurosurgical services in our country.[21,41] Other reasons may include the aforementioned prevailing poor socioeconomic status and/or poor medical insights among these rural dwellers. This may be particularly so in the mild head injured patients and patients with Frankel E spinal cord injury who felt well and saw no need for the requested investigations and outlined management plan and therefore discharged against advice.

DAMA is one of the leading causes of hospital readmission.[12,28,49] This may result from deterioration in the clinical condition of the patients while away from the hospital.[46] Only 4 (4.5%) of our DAMA patients represented to our hospital. In half of these cases, the patients were compelled to do so by family members while the other patients returned to us because of neurological deterioration. One of the patients was a 21-year-old man with Frankel E cervical spine injury who left the hospital without submitting to the requested radiographs. He only represented 18 days after DAMA because of episodes of dizziness and neck stiffness. Cervical spine X-ray showed Type 2 odontoid peg fracture with anterior subluxation of atlas on the axis [Figure 3]. The second was an 18-year-old boy with posterior fossa abscess whose parents DAMA because “the cost of surgery was too much” [Figure 4]. He was brought in dead 4 days after DAMA. One of the patients that were compelled by the relatives to represent had initially been seen at another neurosurgical service before she came to our hospital. She left the first hospital because the relatives felt that the particular hospital was too expensive and came to shop in our hospital for cheaper fees. She represented to our hospital about 2 h after initially signing the DAMA form. Sadly, she discharged against medical advice yet again within 12 h of representation.

Figure 3:

Pre (a) and post (b) reduction cervical spine X-rays of the patient with odontoid peg fracture.

Figure 4:

Axial (a) and sagittal (b) postcontrast cranial computed tomography scan images of the patient with cerebellar abscess.

The negative impact of the privately funded model of payment for health care on the outcome of that care, especially as obtains in the low resource settings of many sub-Saharan Africa has been documented. The absence of health insurance covers in all our patients, and the finding of financial constraints as the most common reason for DAMA lends credence to the need to do away with this model of health-care financing. We advocate alternative health-care financing models particularly a functional and sustainable health insurance scheme in other to maximize the health resources available in these settings.[4]

CONCLUSION

The rate of DAMA was high in our study possibly due to the expensive nature of neurosurgical in-hospital evaluation and treatment in a resource-constrained developing country milieu of privately funded health-care financing. Low income and level of health literacy were the most common causes of DAMA in this neurosurgical patient population. A diagnosis of neurotrauma significantly predicts early DAMA in this study. Functional health insurance scheme with a wide coverage will go a long way in reducing the incidence of DAMA and improving overall health indices in developing countries like ours.