. 2023 Feb 20;177(4):373–383. doi: 10.1001/jamapediatrics.2022.5975

Table 1. Study Characteristics and Screening Metrics for M-CHAT(-R/F).

Source	Total No.^a	Screen age, mo^b	Evaluation age, mo^b	HL, LL, or mixed	M-CHAT language	FU	ASD detection^c	TP	FN	FP	TN	Sens	Spec^d
Baduel et al,¹⁷ 2017	1250	22-26	24-34	LL	French	Y	C: FN strategy	12	6	8	1201	0.667	0.993
Beacham et al,¹⁸ 2018	154	16-45	16-45	HL	English	N	C: all eval	105	19	14	16	0.847	0.533
Canal-Bedia et al,¹⁹ 2011	2480	18-36	18-48	Mixed^e	Spanish	Y	C: weak	23	0	43	2414	1.000	0.980^f
Carbone et al,⁸ 2020	26 364	16-30	46.8 (17.7)^g	LL	English	Y/N^h	P: record	125	253	579	25 407	0.331	0.978
Chang et al,²⁰ 2021	990	17-37	17-37	LL	English	Y	C: FN strategy	31	7	11	941	0.816	0.988
Charman et al,²¹ 2016	543	18-56	32-73	HL	English	N	C: all eval	45	10	32.5	32.5	0.818	0.500
Chlebowski et al,²² 2013	18 989	16-30	25.8 (4.5)	LL	English or Spanish	Y	C: FN strategy	92	6	79	18 289	0.939	0.996
Choueiri et al,²³ 2021	80	18-36	18-36	HL	English	N	C: all eval	53	3	0	24	0.946	1.000
Christopher et al,²⁴ 2021	290	18-48	18-48	HL	English	Y	C: all eval	170	48	48	24	0.780	0.333
Coelho-Medeiros et al,²⁵ 2019	120	16-30	16-30	Mixed^e	Spanish (Chile)	Y	C: FN strategy	18	0	4	97	1.000	0.960
Dereu et al,²⁶ 2012	199	16-31	13-51	HL	Dutch	N	C: FN strategy	10	4	22	163	0.714ⁱ	0.881
DiGuiseppi et al,²⁷ 2010	85	20-86	20-86	HL	English or Spanish	N	C: FN strategy	9	2	26	38	0.818	0.594
Dudova et al,²⁸ 2014	157	≈24	NA	HL	Czech^j	N	C: FN strategy	9	4	9	112	0.692	0.926
Eaves et al,²⁹ 2006	84	17-48	22-53	HL	English	N	C: all eval	48	4	22	8	0.923	0.267
Guo et al,³⁰ 2019	7928	16-30	23 (4)	LL	Chinese	Y	C: FN strategy	72	10	103	7166	0.878	0.986
Guthrie et al,⁹ 2019^k	20 375	16-26	17-88	LL	English or Spanish	Y/N^h	P: record	225	229	1247	18 674	0.496	0.937
Harris et al,³¹ 2021^l	360	24-48	NA	LL	English or Spanish	Y	C: FN strategy	8	29	4	315	0.216	0.987
Hoang et al,³² 2019	17 277	18-30	18-30	LL	Vietnamese	N	C: FN strategy	129	1	118	17 021	0.992	0.993
Inada et al,³³ 2011	1187	17-23	35-44	LL	Japanese	N	P: FN strategy	11	9	46	1121	0.550	0.961
Jonsdottir et al,³⁴ 2022	1586	31.7 (1.7)	NA^m	LL	Icelandic	Y	P: record	18	11	7	1549	0.621	0.988
Kamio et al,³⁵ 2014	1851	17-26	33-73	LL	Japanese	Y	P: FN strategy	29ⁿ	22	24	1661	0.569	0.986
Kanne et al,³⁶ 2018	158	18-48	24-32	HL	English	Y	C: all eval	96	23	24	15	0.807	0.385
Kara et al,³⁷ 2014	618	18-36	24-42	Mixed^e	Turkish	Y	C: FN strategy	45	2	15	534	0.957	0.973
Keehn et al,³⁸ 2021	605	18-48	18-48	HL	English	Y	C: all eval	198	31	234	142	0.865	0.378
Kerub et al,³⁹ 2020	1591	18-36	NA^o	LL	Hebrew	Y	P: record	7	3	43	1538	0.700	0.973
Kim et al,⁴⁰ 2016	827	14-43^p	110-151	HL	English	N	P: all eval	30	28	123	646	0.517	0.840
Kleinman et al,⁴¹ 2008^q	1416	16-30	52.2 (8.0)^g	Mixed^e	English	Y	P: FN strategy	73	7	51	1285	0.913	0.962
Koh et al,⁴² 2014	580	17-48	18-69	HL	English or Chinese	N	C: all eval	158	41	123	258	0.794	0.677
Magán-Maganto et al,⁴³ 2020	6625	14-36	23-36	LL	Spanish	Y	C: FN strategy	15	4	24	6542	0.789	0.996
Matson et al,⁴⁴ 2013	552	16-30	16-30	HL	English	N	C: all eval	150	101	150	151	0.598	0.502
Oner and Munir,⁴⁵ 2020	6712	16-36	16-41	LL	Turkish	Y	C: weak	57	0	95	6388	1.000	0.985
Robins et al,⁴⁶ 2014	16 041	16-30	26.2 (5.5)	LL	English	Y	C: FN strategy	105	18	116	15 496	0.854	0.993
Salim et al,⁴⁷ 2020	143	18-48	18-48	HL	Indonesian	N	C: all eval	16	1	27	99	0.941	0.786
Salisbury et al,⁴⁸ 2018	485	16-48	16-48	HL	English	N	C: all eval	220	77	82	106	0.741	0.564
Samadi and McConkey,⁴⁹ 2015	2941	24-60	24-60	LL	Kurdish or Persian	N	C: FN strategy	25	3	45	2380	0.903^r	0.981
Schjølberg et al,⁵⁰ 2022^s	54 436	19.0 (1.2)	≈42	LL	Norwegian	N	P: record	105	232	4048	50 078	0.312	0.925
Smith et al,⁵¹ 2013	217	18-48	18-48	HL	English	N	C: all eval	97	39	31	50	0.713	0.617
Snow and Lecavalier,⁵² 2008	56	18-48	18-48	HL	English	N	C: all eval	38	5	8	5	0.884	0.385
Srisinghasongkram et al,⁵³ 2016 (sample 1)	109	18-48	18-48	HL	Thai	Y	C: all eval	40	5	1	63	0.889	0.984
Srisinghasongkram et al,⁵³ 2016 (sample 2)	732	18-48	18-48	LL	Thai	Y	C: FN strategy	9	0	1	722	1.000	0.999
Sturner et al,⁵⁴ 2016	5071	18-24	14-41	LL	English	Y	C: FN strategy	23	16	17	4772	0.590	0.996
Sturner et al,⁵⁵ 2022	408	16-20	20.5 (1.9)	LL	English	Y	C: FN strategy	46	17	118	227	0.730	0.658
Taylor et al,⁵⁶ 2014	145	28.1 (4.8)	28.1 (4.8)	HL	English	N	C: all eval	74	12	26	33	0.860	0.559
Toh et al,⁵⁷ 2018	19 297	15-36	NA	LL	Malay, Chinese, or English	N	P: record	18	32	20	19 227	0.360	0.999
Tsai et al,⁵⁸ 2019	317	16-32	36-37	Mixed	Mandarin Chinese	Y	P: all eval	22	3	19	273	0.860^r	0.935
Thi Vui et al,⁵⁹ 2022	40 243	18-30	NA	LL	Vietnamese	N	C: FN strategy	302	3	193	39 726	0.990	0.995
Weitlauf et al,⁶⁰ 2015	74	16-21	18-43	HL	English	Y	C: FN strategy	21	6	7	29	0.778	0.806
Wieckowski et al,⁶¹ 2021^t	3052	17-22	18-60	LL	English or Spanish	Y	C: FN strategy	61	13	79	2729	0.824	0.972
Windiani et al,⁶² 2016	110	18-48	18-48	HL	Indonesian	Y	C: all eval	16	2	5	87	0.889	0.946
Wong et al,⁶³ 2018	236	18-47	18-47	HL	Chinese	N	C: all eval	99	14	58	65	0.876	0.528
Zhang et al,⁶⁴ 2022	11 190	18-24	23.1 (4.6)	LL	Chinese	Y	C: FN strategy	33	15^u	56^u	11 056^u	0.688	0.995

Abbreviations: all eval, all participants evaluated; ASD, autism spectrum disorder; C, concurrent; FN, false negative; FP, false positive; FU, M-CHAT(-R/F) Follow-up administration; HL, high likelihood for ASD; LL, low likelihood for ASD; M-CHAT(-R/F), Modified Checklist for Autism in Toddlers and Modified Checklist for Autism in Toddlers, Revised With Follow-up (original and revised versions combined); N, no; NA, not available; P, prospective; record, medical record review; Sens, sensitivity; Spec, specificity; TN, true negative; TP, true positive; Y, yes; Y/N, Y, but not consistently.

^{^a}

Refers to sample who received the M-CHAT(-R/F).

^{^b}

Range reported for entire sample who received M-CHAT(-R/F), when available. If not available, mean (SD) is reported. If mean (SD) was not available, an estimate from the article or from communication with authors is reported.

^{^c}

Strategy used to detect FN cases.

^{^d}

Specificity was recalculated from the raw numbers so that TN cases included presumed TN results (ie, including children who screened negative but were not further evaluated) for consistency across studies, unless noted otherwise. Negative screen results were presumed to be TN unless there was other presented evidence.

^{^e}

Reclassified as low risk for analyses because most participants were low risk.

^{^f}

True negative and specificity taken directly from article and not recalculated due to missing information.

^{^g}

Age of evaluation is for ASD sample only; age for non-ASD sample is unknown.

^{^h}

Yes/no was reclassified for analyses: Carbone et al⁸ was reclassified as N based on few practices using the Follow-up portion of the M-CHAT-R/F, and Guthrie et al⁹ was reclassified as Y even though the Follow-up portion of the M-CHAT-R/F was not always used, given that it was built into the medical record system and intended to be used when indicated.

^ⁱ

Sensitivity differs slightly from that reported in the article because of a focus on M-CHAT(-R/F) only.

^{^j}

Language presumed as not directly reported in the article.

^{^k}

Values were obtained from the main author for repeated screenings and do not match those reported in the article for single screening.

^{^l}

Subsample of children screened before 48 months of age only is reported. Values were obtained from communication with the main author.

^{^m}

Age of evaluation was up to 18 months after the age of screening.

^ⁿ

P value differs from the one reported in the article because of the addition of 9 nonresponders who needed the Follow-up portion of the M-CHAT-R/F but did not complete it and had confirmed ASD.

^{^o}

Age of evaluation was within 10 months of screenings.

^{^p}

Screen age reported is uncorrected for prematurity.

^{^q}

Study 2 sample only presented and analyzed because of overlap of sample 1 with Chlebowski et al.²²

^{^r}

Sensitivity calculation slightly differs from article’s reported sensitivity owing to rounding in calculation from raw numbers.

^{^s}

Information for M-CHAT²³ is reported.

^{^t}

Information is reported for 18-month screening start age only.

^{^u}

Numbers adjusted to include the 12 screen-negative children with a diagnosis of ASD during subsequent well-child visit and follow-up.