Table 6.

Quality rating as a percentage for 103 articles/reports

Indicator	A (%)	B (%)	C (%)	D (%)
Key cost components	96	0	4	0
Bias	48	49	4	0
Precision	4	20	28	48
Reporting	38	49	14	0

Key Cost Components (note under GHCC nomenclature is Bias Low), a study would receive an ‘A’ if they appropriately account for and report key cost components such as above-service delivery cost, overhead costs, personnel inefficiency/downtime adjustment and value volunteer time. This provides a signal for the completeness of the cost estimated. For the second category, Bias (note under GHCC nomenclature is Bias High), an ‘A’ is given if the study appropriately annuitises capital costs (which is deemed appropriate depending on the stated time since the programme started) and omits unrelated costs, for instance unrelated research costs. These are not the only sources of bias in costing studies, but as with the key cost components, these indicators are considered a signal of potential bias. For the third category, Precision, an ‘A’ is indicated if the study follows sampling, data collection, and reports the cost estimate in a way that correctly reflects the level of precision of the study. These include sampling at a country or site level as appropriate, selecting and reporting on a relevant cost allocation method, resource identification, the method of measuring output, and the number of sites selected. The fourth category is Reporting. An ‘A’ would be given for this category if the authors explicitly report key methods and results, such as the urbanicity (rural/urban/peri-urban or a mixture), the ownership (public, NGO etc.), intervention components and breakdown by activity. In all the categories outlined above, completely omitting or failing to account for these aspects would result in a ‘D’. Being given a low score for Key Cost Components could indicate an underestimation of unit costs, while a low score for Bias could indicate an overestimation