Challenges and experiences in linking community level reported out-of-pocket health expenditures to health provider recorded health expenditures: Experience from the iHOPE project in Northern Ghana.

Out of pocket health payment (OOPs) has been identified by the System of Health Accounts (SHA) as the largest source of health care financing in most low and middle-income countries. This means that most low and middle-income countries will rely on user fees and co-payments to generate revenue, rationalize the use of services, contain health systems costs or improve health system efficiency and service quality. However, the accurate measurement of OOPs has been challenged by several limitations which are attributed to both sampling and non-sampling errors when OOPs are estimated from household surveys, the primary source of information in LICs and LMICs. The incorrect measurement of OOP health payments can undermine the credibility of current health spending estimates, an otherwise important indicator for tracking UHC, hence there is the need to address these limitations and improve the measurement of OOPs. In an attempt to improve the measurement of OOPs in surveys, the INDEPTH-Network Household out-of-pocket expenditure project (iHOPE) developed new modules on household health utilization and expenditure by repurposing the existing Ghana Living Standards Survey instrument and validating these new tools with a 'gold standard' (provider data) with the aim of proposing alternative approaches capable of producing reliable data for estimating OOPs in the context of National Health Accounts and for the purpose of monitoring financial protection in health. This paper reports on the challenges and opportunities in using and linking household reported out-of-pocket health expenditures to their corresponding provider records for the purpose of validating household reported out-of-pocket health expenditure in the iHOPE project.

Effect of specificity of health expenditure questions in the measurement of out-of-pocket health expenditure: evidence from field experimental study in Ghana.

BACKGROUND: The effect of number of health items on out-of-pockets (OOPs) has been identified as a source of bias in measuring OOPs. Evidence comes mostly from cross-sectional comparison of different survey instruments to collect data on OOPs. Very few studies have attempted to validate these questionnaires, or distinguish bias arising from the comprehensiveness of the OOPs list versus specificity of OOPs questions. OBJECTIVES: This study aims to estimate biases arising from the specificity of OOPs questions by comparing provider and household's information. METHODS: A generic questionnaire to collect data on household's OOPs was developed following the nomenclature proposed in division 6 of the classification of household final consumption 2018. The four categories within such division are used to set the comprehensiveness of the OOPs list, the specificity within each category was tailored to the design of the nationally representative living standard survey in Ghana where a field experiment was conducted to test the validity of different versions. Households were randomised to 11, 44 or 56 health items. Using data from provider records as the gold standard, we compared the mean positive OOPs, and estimated the mean ratio and variability in the ratio of household expenditures to provider data for the individual households using the Bland-Altman method of assessing agreement. FINDINGS: We found evidence of a difference in the overall mean ratio in the specificity for OOPs in inpatient care and medications. Within each of these two categories, a more detailed disaggregation yielded lower OOPs estimates than less detailed ones. The level of agreement between household and provider OOPs also decreased with increasing specificity of health items. CONCLUSION: Our findings suggest that, for inpatient care and medications, systematically decomposing OOPs categories into finer subclasses tend to produce lower OOPs estimates. Less detailed items produced more accurate and reliable OOPs estimates in the context of a rural setting.

Investigating the effect of recall period on estimates of inpatient out-of-pocket expenditure from household surveys in Vietnam.

Out-of-pocket payments (OOPs), direct payments by households or individuals for healthcare are part of the health financing landscape. Data on OOPs is needed to monitor progress in financial risk protection, and the evaluation of health financing policies. In low-and-middle-income countries, estimates of OOPs rely heavily on self-reported data from household surveys. These surveys require respondents to recall events in the past and can suffer from recall biases. This study investigates the effect of recall period on the agreement of the amount and timing of inpatient OOPs between household reports and provider records in Bavi, Vietnam. We recruited 1397 households for interview using records from the district hospital. The households were interviewed with identical questionnaires except that the recall period was either 12 or 6 months. We linked household with provider data and excluded medicine costs from both household and provider OOPs since they could be purchased outside the hospital. We estimated the effect of recall period on the overall mean and variability of ratios of household to hospital reported OOPs using the Bland-Altman approach for method comparison. We estimated the effect of recall period on whether a transaction was recalled correctly in expenditure and time using multinomial regression. The households reported higher amounts of OOPs than did the hospital for both recall periods. There was no evidence of an effect of recall period on the mean of the ratios of household- to hospital-reported OOPs, although the confidence intervals are not inconsistent with previous studies indicating higher OOPs for shorter recall periods. The geometric mean ratio for the 6-month period was estimated to be a multiple of 1.4 (95% CI 0.9, 2.1) times that of the 12-month period. Similarly, there was no evidence of an effect of recall period on the risk of reporting lower or higher amounts than provider OOPs. The occurrence and timing of inpatient stays generally recalled well, with 70% remembered in the correct month declining slightly over time. Respondents for the 6-month recall period had a significantly lower risk of failing to report the event (RR 0.8 (0.7, 1.0)). The results suggest the best recall period may depend on whether the purpose of a survey is for the recall of the timing of events, in which case the 6 month period may be better, or the amounts of OOPs, where there was no significant difference and the provider records are not a gold standard but the 12 month period had a tendency to be in closer agreement with the provider OOPs.

Socio-demographic determinants of low birth weight: Evidence from the Kassena-Nankana districts of the Upper East Region of Ghana.

OBJECTIVE: To examine the social, economic and demographic factors that determine low birth weight in the two Kassena Nankana districts of the Upper East region of Ghana. METHODS: Cross-sectional data was collected from January 2009 to December 2011 using the Navrongo Health and Demographic Surveillance System which monitors routine health and demographic outcomes in the study area. Data on foetal characteristics such as birth weight, and sex and maternal age, parity, maternal education, marital status, ethnicity, religious affiliation and socio-economic characteristics were collected and described. Tests of means, proportions and Chi-squares are employed in bivariate analysis, and adjusted logistic regression models fitted to control for potential confounding variables. All tests were two-sided and test of significance was set at p-value of < 0.05. RESULTS: There were 8,263 live births (44.9% females) with an overall average birth weight of 2.85 kg (2.9 kg for males and 2.8 kg for females). The average maternal age was 28 years, median parity 2, maternal literacy rate was about 70% and 83% of mothers were married. The prevalence of low birth weight was 13.8% 95%CI [13.10, 14.6] and more in female babies than in males (15.5% vs 12.2%; p<0.0001). Determinants of low birth-weight after controlling for confounding factors were sex of neonate (OR = 1.32, 95%CI [1.14,1.52]; p<0.0001), maternal age (p = 0.004), and mothers who are not married (OR = 1.44 [1.19, 1.74]; p<0.0001). CONCLUSION: Female neonates in this population were likely to present with low birth weight and maternal factors such as younger age, lower socio-economic status and single parenthood were major determinants of low birth weight. Effective and adequate antenatal care should therefore target women with these risk factors.