Table 4.

Comparison of multivariable-adjusted changes^a in percentage of providers with positive perceptions of EHR benefit

Domain	Individual Items	Practice Setting		Age		Specialty
		Inpatient Only	Any Outpatient	<= 45	> 45	Primary Care	Medical Specialty	Surgical Specialty
Overall satisfaction	Overall satisfaction	–	↓	–	↓	–	–	–
Productivity	Productivity	–	↓	↓	↓	–	↓	↓
Overall patient safety	Overall patient safety	–	–	–	–	–	–	–
Better care	Better patient care	–	↓	↓	↓	–	↓	↓
Medical record completion burden	Charting after hours	–	–	–	–	–	–	–
Patient safety and quality	Clinical decision quality	–	↓	–	↓	–	↓	↓
	Avoid drug errors	–	–	–	–	–	–	–
	Avoid drug allergy	–	–	–	–	–	–	–
	Avoid drug interaction	–	–	–	–	–	–	–
	Critical lab value alert	–	–	–	–	–	–	–
Communication	Provider communication	–	–	–	–	–	–	–
	Patient communication	–	↑	↑	–	↑	–	–
Information exchange	Hospital transition	–	↑	↑	–	↑	–	–
	Easy access	–	↓	–	↓	–	↓	–
	Timely information	–	↓	–	↓	–	–	–
Preventive care	Preventive care delivery	–	↑	–	–	–	–	–
	Preventive care prompt	–	↑	–	↑	↑	–	–
	Preventive lab prompt	–	–	–	–	–	–	–
Clinical workflow	Scheduling consults	–	–	–	–	–	–	–
	Receiving consult reports	–	–	–	–	–	–	–
	Monitoring patients	–	↓	–	↓	–	–	–
	More time for patients	–	↓	–	↓	–	↓	–
	Coordination of care	–	↓	–	↓	–	↓	–
	Scheduling patients	–	–	–	–	–	–	–
	Drug refills	–	–	–	–	–	–	–
	Access to records	–	–	–	–	–	–	–
Technical support and access	Computer access	–	–	–	–	–	–	–
	Adequate resources	–	–	–	↓	–	–	–
General satisfaction	Satisfaction with ease of use	–	↓	–	↓	↓	↓	–
	Satisfaction with reliability of system	–	↑	–	↑	↑	–	–
	Satisfaction with sharing medical information	–	–	–	↑	–	–	–
	Satisfaction with obtaining medical information	–	–	–	–	–	–	–

^aSignificant changes in the percentage of providers with positive perceptions from baseline to long-term follow-up (eg, % satisfied, % agree) based on multivariate-adjusted analyses. Change in each item was adjusted for age, gender, provider training (physician vs other provider), setting (inpatient vs any outpatient), time using electronic health records. P-values were also adjusted for multiple comparisons using the Benjamini-Hochberg method.¹⁴

Increase ↑, no significant change –, decrease ↓ from baseline to long-term follow-up.