Addressing health disparities using multiply imputed injury surveillance data

Table 1 Frequency analysis of missingness: unweighted counts and percentages for missing variables in NEISS-AIP^a, United States, 2014–2018

Missing Variables^b	2014		2015		2016		2017		2018		2014–2018
Missing Variables^b	n	%	n	%	n	%	n	%	n	%	N	%
LOC	227,655	36.8	215,807	35.2	226,014	36.4	220,506	34.5	193,958	33.4	1,083,940	35.3
RACE	184,732	29.8	210,518	34.4	211,600	34.0	209,781	32.8	188,437	32.4	1,005,068	32.7
CAUSE	17,924	2.9	16,897	2.8	18,557	3.0	19,031	3.0	16,048	2.8	88,457	2.9
BDYPT	7,386	1.2	8,306	1.4	8,876	1.4	8,691	1.4	9,862	1.7	43,121	1.4
TYPE	5,507	0.9	5,359	0.9	4,861	0.8	5,468	0.9	5,450	0.9	26,645	0.9
AGE	199	0.03	254	0.04	251	0.04	238	0.04	198	0.03	1,140	0.04
DISP	16	0.003	6	0.001	7	0.001	204	0.03	15	0.003	248	0.008
SEX	8	0.001	13	0.002	7	0.001	22	0.003	23	0.004	73	0.002
Missing Total ^c	346,357	55.9	349,875	57.1	366,692	59.0	368,159	57.6	324,396	55.8	1,755,479	57.1
Non-missing Total ^d	272,986	44.1	262,545	42.9	254,880	41.0	271,355	42.4	256,529	44.2	1,318,295	42.9
Total Observations	619,343		612,420		621,572		639,514		580,925		3,073,774

^a NEISS-AIP National Electronic Injury Surveillance System-All Injury Program
^b LOC location where the injury occurred, RACE race and ethnicity, CAUSE external cause of injury, BDYPT primary body part affected, TYPE work-relatedness, AGE age in years, DISP disposition at emergency department discharge, SEX gender
^cMissing Total: the cumulative missing where observations had at least one missing variable
^dNon-missing Total: observations had no missing values

ISSN: 1475-9276