Development, validation, and usage of metrics to evaluate the quality of clinical research hypotheses

Table 2 Example of hypotheses and evaluation results using the 3-item instrument

Hypothesis	Three evaluation items	R1	R2	R3	R4	R5	R6	R7	Item means	SD	Hypothesis mean ± 95% confidence interval
H1	Validity	5	5	5	3	4	3	4	4.14	0.83	4.0 ± 0.35
	Significance	4	4	4	3	3	2	4	3.43	0.73
	Feasibility	5	4	4	4	5	5	4	4.43	0.49
H2	Validity	NA	3	4	3	3	3	1	2.83	0.90	2.64 ± 0.52
	Significance	NA	3	4	4	2	3	2	3	0.96
	Feasibility	NA	3	2	3	2	1	1	2	0.82

Note: Hypothesis 1: Patients who have hypertension between 2005 and 2015, do hypertension patients have a higher obesity morbidity rate (ICD9 codes: 27801) in 2015 than in 2005? Hypothesis 2: Whether the changes in packed food consumption caused an increase in diabetes (ICD9 code: #250) from 2005 (case counts: 774) to 2015 (case counts: 1281) at the zip code level? R1: reviewer 1; NA: not applicable (i.e., an evaluator cannot assess the item, a hypothesis is invalid, and all following items are not evaluated)

ISSN: 1471-2288