Financial implications of preterm birth during initial hospitalization: The extent and predictors of catastrophic health expenditure

Preterm birth incidence has risen globally and the high cost of initial hospitalization poses financial burden to the family. This study assessed family cost at neonatal intensive care units of two hospitals in the state of Kedah, Malaysia. Family’s expenditure was obtained using a structured questionnaire. 126 families who were government employed spent a mean total cost of MYR 549 (MYR 0 - MYR 4,700) compared to MYR 650 (MYR 40 – MYR 9,300) for 244 families who were not government employed. Mean income loss was MYR 310 (MYR 0 – MYR 15,000) and MYR 348 (MYR 0 – MYR 5,500) respectively. Travel expenses was the cost driver for all families. 15% of families in this study were already living below the income poverty line and majority were not government employed. For the rest of the families, 21% became impoverished when one month household income was used for hospitalization cost but this lowered to 9% with cumulative household income by length of hospital stay. Overall incidence of catastrophic health expenditure among families was 38%. Using multivariable logistic regression household income and residential location were predictive factors for catastrophic health expenditure. Despite universal health coverage through subsidy of direct medical (hospital) cost, the high incidence of catastrophic health expenditure and impoverishment among families of preterm infants was attributable to out of pocket payment for direct non-medical cost (such as travel and food) and indirect cost from income loss. Government employed families with an array of employment benefits appear better protected against financial hardship compared to those in private sector or self-employed. Remedial measures include improving neonatal intensive care unit rooming-in service for mothers, complementary financial assistance for families and enhancing universal health coverage through affordable social health insurance for infant healthcare.


35
Preterm birth is defined as delivery before 37 completed weeks of gestation. Preterm birth is 36 increasingly common with substantial medical, economic and social impact as it is invariably 37 associated with acute and chronic complications (1, 2). Since its inception in 2009, Malaysia's 38 preterm birth registry showed an increasing rate from 8.1% to 11.3% between 2010 and 2012 (3).

39
Due to advancements in care over the last few decades, outcome and survival of preterm infants 40 have improved, however, the economic impact of preterm care has gained much attention (1,2). 41 Most studies have been devoted to cost of intensive care as initial hospitalization accounted for the 42 bulk of health care cost during the first 2 years of life of a preterm infant (4). However, cost analysis 43 during initial hospitalization had rarely taken into account the family's perspective (2) 44 Preterm birth has been found to cause significant out of pocket (OOP) spending for families of 45 preterm infants with estimated cost of up to 2% to 4% of gross annual income during neonatal 46 period (2,5). For extremely low birth weight infants it was estimated that parental mean cost prior to discharge was up to 4% of the total cost (4). Travel expenses contributed 64% of these nonreimbursable payments, 30% from loss of earnings and 6% from accommodation during visits.

49
Compounding the situation is the fact that OOP payment is the most prevalent means of healthcare 50 payment in Asian countries and households are at risk of catastrophic health expenditure (CHE) and job schedule and lost wages (if paid by hour or day) and missed working days from time spent for 104 hospital visits that otherwise would have been spent for working. Indirect cost or productive work 105 time lost was assessed using the human capital approach where productive work time lost refers to 106 the work output that would have been generated if the illness event had not occurred (10,11

178
Respondents' demography 179 The response rate among eligible families in this study was 100%. The majority of families of preterm 180 infants were Malays (89%) with Indian (5%), Chinese (4%) and others (2%) making up the ethnic 181 groups ( Table 1). 80% of fathers and 90% of mothers were aged between 20-40 years. 1% of fathers 182 and 4% of mothers were below the age of 20. Parents with secondary education were the biggest 183 group (58% respectively for paternal and maternal groups), followed by tertiary education (34% and 184 36% respectively) and primary level or lower (less than 10% in each group). Only 24% of fathers 185 worked in the government sector while the rest were either self-employed (36%) or in the private 186 sector (40%). Unsurprisingly majority of mothers were homemakers (54%) with the rest in 187 government service (22%), private sector (18%) and self-employed (6%). There were more families 188 of preterm infants who came from outside the hospital district (58%) compared to within the district (42%).

Financial sources
319 Approximately a third of families in this study had the benefit of full government subsidy to cope 320 with hospital cost (direct medical cost). The remaining families who received partial government 321 subsidy resorted to OOP payment for the remainder of hospital cost. Various strategies were used in 322 this group but majority (one fourth) of them opted for current income of any household member 323 followed by savings and borrowing from friends or relatives. No family in this study had the option of 324 insurance coverage for hospitalization of their preterm infants. These findings are in line with the 325 fact that for most of Asian population OOP payment is the primary method for health care financing 326 (6, 7). A study in neighbouring Indonesia revealed borrowing as a dominant coping strategy among 327 the poor to acquire health care during illness (18). In this study frequency of borrowing or selling assets for health care payment was lower than 26% noted in a study of lower and middle income received little financial assistance due to limited and discretionary funds. A similar situation of 333 undisclosed financial difficulties may exist in this study where more than two thirds of non-334 government employed parents perceived hospital cost payment as 'not difficult at all' or 'not that 335 difficult' and less than a third found it to be 'difficult' or 'very difficult'. This was despite the fact that 336 more than a fifth of them were living below the poverty line, more than half suffered income loss 337 and close to a fifth were unable to make full payment of the partially subsidised amount upon 338 discharge. Despite full government subsidy of hospital cost government employed parents received 339 full too experienced income loss, impoverishment and CHE but at a comparatively much lower rate 340 than their counterparts.

341
Overall, 15% of families this study were living below the income poverty line and the overwhelming 342 majority was from the non-government employed group. This was surprisingly very much higher 343 than the 0.6% incidence of poverty for Malaysia and 0.3% for the state of Kedah in 2014 (14). With 344 exclusion of families below poverty line further analysis was done and it was found that up to a fifth 345 of families above poverty line suffered impoverishment when one month of HHI was used for initial 346 hospitalization cost but this improved to less than a tenth when cumulative HHI (according to LOS) 347 was used. Subsequently analysis was performed on government and non-government employment 348 status to assess the impact of full and partial subsidy of hospital cost on impoverishment.