Table 1.

Regular expressions used in final rule-based natural language processing algorithm.

Information searched	Regular expressions * used in query	Examples
Sexual history	sex[a-z,\.,\s,\/]|intercourse	MATCH: “sexual activity,” “does not have sex,” no sex/drug use,” “had sex” NOT A MATCH: “Sex: Male”
Protection used	protect|condom condom cath|child protect	MATCH: “uses protection,” “uses condoms some of the time” NOT A MATCH: “condom catheter,” “child protection services”
STI testing offered/previously performed	GC[∧S]|G\/C|gonorr| gonnor|gonor|gonoc|chlam| ST[I|D]\s(test|screen|lab)| test(ing|ed|\s)\sfor\sST[I|D] chlam pneumo|chlamydia pneumo	MATCH: “will send GC/CT,” “STD testing,” “test for STI,” “chlamydia” NOT A MATCH: “chlamydia pneumonia”
Not sexually active	denies (any sex|history of sex|hx of sex|intercourse|hx of intercourse|being sexual|ever being sexual)|no sex|never (had sex|been sexual)|not sexually active	MATCHES: “denies ever being sexually active,” “never been sexually active” NOT A MATCH: “denies ever having been sexually active”
Partner gender	Within substring of text 50 characters prior to and 50 characters after/ sex[a-z,\.,\s,\/]|intercourse/ male/boy/girl/men/man	MATCH: “has sex with female partners,” “hx of sexual activity with women” NOT A MATCH: “interested in men” (with no mention of sex or intercourse within 50 characters)
Type of sexual practice	(oral|vaginal|anal)\s(sex|penetration| intercourse)	MATCH: “has had oral sex,” “anal penetration” NOT A MATCH: “has had sex—both oral and anal”
Contraception used	OCP|contraceptive|contraception, birth control|planon|IUD|nuvaring| depo	MATCH: “is on OCP,” “Depo-provera 1 month ago,” “on birth control for PCOS” NOT A MATCH: “on mini pill”

^*All regular expressions are given in Python format and were multi-line and case insensitive.