. 2019 Jun 10;34(11):2355–2367. doi: 10.1007/s11606-019-05025-3

Table 5.

Variation of Documentation Categories

Variation category	Documentation category	Completion (median across physicians)	Interquartile Range (25%‘ile–75%‘ile)	Percent of explained variation at each level			Remaining variation at the physician level
Variation category	Documentation category	Completion (median across physicians)	Interquartile Range (25%‘ile–75%‘ile)	State	Provider Organization	Practice	Remaining variation at the physician level
Low	Sign-off/close encounter	100%	0%
Low	Sign-off/close encounter	100%	100–100%
Low	Creating/sending out orders	100%	0.7%
Low	Creating/sending out orders	100%	93.3–100%
Low	Collect vitals	99.1%	4.1%
Low	Collect vitals	99.1%	95.9–100%
Low	Collect clinical encounter reason	98.2%	7.1%
Low	Collect clinical encounter reason	98.2%	92.9–100%
Low	Conduct physical exam	96.6%	9.0%
Low	Conduct physical exam	96.6%	90.1–99.1%
Low	History of present illness	95.2%	13.8%
Low	History of present illness	95.2%	85.5–99.3%
Low	Medication list	91.5%	19.5%	7.7%	20.7%	27.8%	43.7%
Low	Medication list	91.5%	77.9–97.5%	[0.15]	[0.15]	[0.15]	[0.46]
High	Social history	76.1%	53.3%	4%	17.8%^***	16%^**	62.2%^***
High	Social history	76.1%	39.5–92.8%	[0.02]	[0.04]	[0.05]	[0.03]
High	Conduct review of systems	73.5%	62.3%	3.2%	15.6%^**	13.5%^*	67.7%^***
High	Conduct review of systems	73.5%	32.9–95.2%	[0.03]	[0.05]	[0.05]	[0.04]
High	Problem list	33.7%	73.1%	1.3%	9.7%^*	19%^**	70.1%^***
High	Problem list	33.7%	3.5–76.6%	[0.01]	[0.04]	[0.06]	[0.04]
High	Review or discuss documents	18.7%	50.8%	7.1%*	5%	9.9%	78.1%***
High	Review or discuss documents	18.7%	10.3–61.2%	[0.03]	[0.04]	[0.06]	[0.04]
High	Assessment and diagnosis	13.4%	60.4%	0%	8.3%	15.8%*	76%***
High	Assessment and diagnosis	13.4%	2.6–62.9%	[0.01]	[0.04]	[0.06]	[0.05]
Low	Interpret incoming clinical documents	14.6%	17.3%	2%	15.3%**	37.7%***	45%***
Low	Interpret incoming clinical documents	14.6%	8.1–25.4%	[0.01]	[0.05]	[0.06]	[0.04]
Low	Conduct procedure	3.7%	6.7%
Low	Conduct procedure	3.7%	1.8–8.5%
Low	Confidential information	2.6%	5.2%
Low	Confidential information	2.6%	1.1–6.3%

Note: Bootstrapped standard errors appear in brackets. Because models did not converge for most low-variation tasks, we only include regression results for high-variation tasks and two low-variation tasks. With the exception of creating and sending out orders, which was at minimum completed in 14% of encounters, and interpreting incoming clinical documents, which was at maximum completed in 73% of encounters, the range of documentation completion was 0–100%

*p significant at the 0.05 false discovery rate; **p significant at the 0.01 false discovery rate; ***p significant at the 0.001 false discovery rate