Personal Health Behaviors are Associated with Physical and Mental Unhealthy Days: A Prescription for Health (P4H) Practice-based Research Networks Study

Desireé B. Froshaug; L. Miriam Dickinson; Douglas H. Fernald; Larry A. Green

doi:10.3122/jabfm.2009.04.080150

Article Figures & Data

Tables

View popup

Table 1.

Comparison of Patient Characteristics for Those Included Versus Those Excluded²⁰

Patient Characteristics	Included in Analysis (n = 3738)	Excluded from Analysis (n = 1620)	P	Total (n = 5358)
Risk behaviors
Current smokers (n [%])	920 (24.61)	370 (25.22)	.6468	1290 (24.78)
Diet score (mean [SD])	5.66 (2.57)	5.58 (2.54)	.2854	5.64 (2.56)
MET-minutes (mean [SD])	2356 (3202)	2145 (2916)	.0407	2308 (3141)
Gender (n [%])			.0009
Men	1010 (27.02)	498 (31.50)		1508 (28.35)
Women	2728 (72.98)	1083 (68.50)		3811 (71.65)
Age (years) (mean [SD])	48 (16.00)	55 (17.00)	<.0001	50 (16.60)
Race/ethnicity (n [%])			.1889
Non-Hispanic white	2229 (61.09)	883 (58.21)		3112 (60.24)
Non-Hispanic black	612 (16.77)	278 (18.33)		890 (17.23)
Hispanic/Latino	631 (17.29)	287 (18.92)		918 (17.77)
Other	177 (4.85)	69 (4.55)		246 (4.76)
Education (n [%])			<.0001
<High school	457 (12.23)	321 (19.81)		778 (14.52)
High school graduate or GED	982 (26.27)	466 (28.77)		1448 (27.03)
>High school	2174 (58.16)	674 (41.60)		2848 (53.15)
No response	125 (3.34)	159 (9.81)		284 (5.30)
Income (per year) (n [%])			<.0001
<$25,000	1373 (36.76)	686 (42.35)		2059 (38.43)
$25,000–$49,999	850 (22.74)	321 (19.81)		1171 (21.86)
≥$50,000	995 (26.62)	359 (22.16)		1354 (25.27)
No response	520 (13.91)	254 (15.68)		774 (14.45)
Body mass index (mean [SD])	31.09 (8.09)	29.98 (7.42)	<.0001	30.76 (7.92)
Overall unhealthy days (n [%])			.4445
0	1017 (27.21)	228 (28.97)		1245 (27.51)
1–13	1432 (38.31)	284 (36.09)		1716 (37.92)
≥14	1289 (34.48)	275 (34.94)		1564 (34.56)
Physically unhealthy days (n [%])			<.0001
0	1586 (42.43)	401 (35.64)		1987 (40.86)
1–13	1384 (37.03)	449 (39.91)		1833 (37.69)
≥14	768 (20.55)	275 (24.44)		1043 (21.45)
Mentally unhealthy days (n [%])			.0999
0	1692 (45.26)	445 (48.11)		1987 (45.83)
1–13	1236 (33.07)	272 (29.41)		1833 (32.34)
≥14	810 (21.67)	208 (22.49)		1043 (21.83)

Bolded values signify P < 0.05.

View popup

Table 2.

The Relationship Between Unhealthy Days and Risk Behaviors: a Multilevel Ordinal Regression Model²⁰ (n = 3738)

	ML Estimate	Standard Error	tTest Score	P	Adjusted Odds Ratio (95% CI)
Threshold 2^*	1.7919	0.0423	42.37	<.0001
Intercept	0.6137	0.1245	4.93	<.0001	1.00 (1.00)
Patient risk behaviors
Smoking	0.4148	0.0759	5.47	<.0001	1.51 (1.30, 1.76)
Diet Score^†^‡	0.0966	0.0128	7.56	<.0001	1.10 (1.07, 1.13)
Exercise MET minutes^‡	-0.0375	0.0102	-3.70	.0005	0.96 (0.94, 0.98)
Patient demographics
Age^‡	-0.0105	0.0022	-4.83	<.0001	0.99 (0.99, 0.99)
Female	0.3367	0.0741	4.55	<.0001	1.40 (1.21, 1.62)
African-American^§	-0.1315	0.0916	-1.44	.1559	0.88 (0.73, 1.05)
Hispanic^§	-0.2318	0.1084	-2.14	.0363	0.79 (0.64, 0.98)
Low income	0.5867	0.0759	7.73	<.0001	1.80 (1.55, 2.09)
Practice characteristics
Registry^‖	-0.2045	0.1142	-1.79	.0782	0.82 (0.65, 1.02)
University owned	0.4389	0.1491	2.94	.0045	1.55 (1.16, 2.08)
Hospital owned	-0.0214	0.1056	-0.20	.8397	0.98 (0.80, 1.20)
Other	0.4849	0.1697	2.86	.0058	1.62 (1.16, 2.26)

Bolded values signify P < 0.05.
* With intercept plus coefficients, model estimates log odds (ML estimate) of having 0 unhealthy days as opposed to 1–30 unhealthy days. The addition of threshold 2 to the log odds estimates log odds of having 0–13 unhealthy days compared to 14–30 unhealthy days.
† Ranging 0–14, where 0 is best and 14 is worst diet.
‡ Centered at the mean of the sample.
§ Compared with non-Hispanic whites.
‖ Compared with private clinician-owned.