How well does the minimum data set measure healthcare use? a validation study

Table 3 Levels of Agreement Between MDS and Administrative Health Care Use Records; Value (95% Confidence Limits)

	Count	Sensitivity	Specificity	Positive Predicted Value	Negative Predicted Value	Kappa
Nursing Home User Status
i) Newly Admitted NH Residents	8695	0.97 (0.96; 0.98)	0.99 (0.99; 1.00)	0.99 (0.98; 0.99)	0.98 (0.98; 0.99)	0.97 (0.96; 0.97)
ii) Residents Admitted from Hospital versus Another Location^a	3164	0.91 (0.90; 0.92)	0.82 (0.80; 0.84)	0.91 (0.90; 0.92)	0.83 (0.81; 0.85)	0.74 (0.71; 0.76)
ii) Residents with 1+ NH Transfer	8695	0.86 (0.84; 0.88)	0.99 (0.99; 0.99)	0.98 (0.97; 0.99)	0.98 (0.97; 0.98)	0.90 (0.89; 0.92)
iii) Residents who Died	8695	0.94 (0.93; 0.95)	0.99 (0.99; 1.00)	0.99 (0.99; 0.99)	0.97 (0.96; 0.97)	0.95 (0.94; 0.95)
Health Care Use
i) Hospital Inpatient Visits^b	20,551	0.58 (0.55; 0.61)	0.97 (0.97; 0.97)	0.45 (0.42; 0.48)	0.98 (0.98; 0.98)	0.49 (0.46; 0.51)
Frequency of visits^c	484	0.60 (0.46; 0.72)	0.90 (0.86; 0.92)	0.41 (0.30; 0.52)	0.95 (0.92; 0.97)	0.41 (0.29; 0.52)
ii) ED Visits not ending in Hospitalization^b	20,551	0.34 (0.32; 0.36)	0.98 (0.98; 0.98)	0.57 (0.54; 0.60)	0.95 (0.94; 0.95)	0.39 (0.37; 0.42)
Frequency of visits^c	532	0.34 (0.26; 0.44)	0.92 (0.90; 0.95)	0.54 (0.42; 0.65)	0.85 (0.81; 0.88)	0.31 (0.21; 0.41)
iii) Days of Physician Examinations	44,003	0.52 (0.52; 0.53)	0.72 (0.71; 0.72)	0.63 (0.63; 0.64)	0.62 (0.61; 0.62)	0.24 (0.23; 0.25)
Frequency of days^c	11,135	0.24 (0.23; 0.26)	0.87 (0.86; 0.88)	0.47 (0.44; 0.49)	0.71 (0.70; 0.72)	0.13 (0.11; 0.15)

^aConducted on residents defined as newly admitted in both data systems
^bResults exclude MDS assessments completed within 90 days of any NH admission (N = 4354)
^cAnalyses conducted on residents with one or more health care contacts in each data system

ISSN: 1472-6963