Quality Care Improvement Program in a Community-Based Participatory Research Project: RESULTS

Posted by James

Provider Retention

Forty-seven healthcare providers practicing alone or in a group in 15 institutions were recruited to participate in the CQIP program (Table 1). All but two providers were physicians. These providers were a physician assistant affiliated with a practice that dropped out at the end of the first year and a nurse practitioner with a practice that stayed three years in the program. No new practice entered the project after it started. However, following expansion of one practice in 1999-2000, four new physicians were enrolled. The biggest drop-out occurred at the end of the first intervention year because of disbandment of three practices with 14 providers that were owned by the same group (the medical director for these practices left and the providers discontinued their participation) and withdrawal of one provider. Thereafter, the drop-out rate remained constant (12-14%), with about 40% of providers still participating at the end of the fourth year. On a year-to-year basis, however, the project objective of retaining 70% of providers was met (83%, 88%, and 82% the last three study years, respectively). It is of interest to note that single practitioners were more likely to stay in the program. The reasons for withdrawing varied. Only one single-practitioner practice truly declined CQIP participation; others left for migration out of Raleigh (n=2 practices with six providers), retirement (two providers in a practice that remained in the program), and practice disbandment or reorganization (n=6 practices with 21 providers).

Practices that remained in the program were located mainly in the southeast Raleigh target area; those that dropped out were located outside the target area, though they saw patients who resided in southeast Raleigh (Table 2). Those who remained in the program also had a greater African-American (range: 44-90% vs. 8.5-50% patient, p<0.0003) and minori ty patient and provider mix and a greater relative diabetic population size than those who dropped out. The majority of institutions were community-based practices (only one out of the 15 practices was an academic institution). In addition, the mean HbAlc at baseline for patients in practices who participated in all four years was higher than for those in practices that dropped out in years two through four (9.9% ± 0.7% vs. 8.9% ± 0.4%, p<0.05). Providers who stayed in the program did not differ from those who left, except for the distribution of specialties (Table 2), and there was no significant difference in the total amount of yearly contacts. The mean number of contacts ± standard deviation increased from 3.1 ±3.5 the first year to 6.7 ±1.9 the fourth year, with a reduction in the third year to 2.3 ± 2.9.
tadalis sx 20

Table 1. Retention of Practices and Providers over Time. Project DIRECT, 1996-2000

	Baseline	Year 1	Year 2	Year3	Year 4
Practices, n	15	12	10	10	6
Single practitioner	5	5	4	4	4
Multiple practitioner	10	7	7	6	2
Healthcare provider, n	47	30	25	22	18*
Y-to-Y retention,* %		64	83	88	82
Overall retention, t %		64	53	53	38
* Y-to-Y: year to year, measure of annual objective achievement; denominator = number of practitioners at the end of previous year; t Measure of intervention overall effectiveness: denominator = number of practitioners at start of intervention; J Five newly recruited healthcare providers in a participating practice.

Patterns of Care Over Time

Data for patterns of care were abstracted from 1,006 charts, of which 410 were collected in practices

that stayed in the program all four years. Characteristics of the population were assessed on the former, whereas prevalence of care was assessed on the latter.

Table 2. Baseline Characteristics of the Practice/Providers by Retention Status. Project DIRECT, 1996-2000

Characteristics                                        Retention Status in CQIP Intervention			P Value*
	Yes	No	meanp = теапр
N practices	6	9
N providers	18	29
Practice-Related
Location: southeast Raleigh vs. not, %	85.7	0.0
Size of the practice: mean number of HCPs	3.1	4.7	0.33
Percentage of providers surveyed	77.4	82.6	0.74
Patient population size and management
Absolute number of diabetic patients/week/provider	36.5	15.7	0.05
Absolute total number of patients/week/provider	113.1	85.4	0.32
Relative diabetic population size/week/per provider!	0.3	0.2	0.05
Patients’ ethnic mix:
% African Americans	66.7	29.0	0.0003
% all minority populations	71.7	34.6	0.0001
Percentage of African-American providers	38.1	16.7	0.30
Percentage of minority providers	52.4	25.0	0.21
Provider-Related
Age	45.3	40.0	0.17
Gender, % male	50.0	52.0	0.90
Years since degree, median (range)	11.5 (2-62)	10 (0-27)	0.52
Providers’ specialty, %
General medicine	14.3	20.0	0.002
Internal medicine	50.0	3.3
Family medicine	35.7	70.0
Psychiatry	0.0	6.7
* MeanR = mean among retained providers; Meanp = mean among providers who dropped out; t
Relative diabetic population size at baseline = (weekly # diabetic patients/weekly # all patients).

Ages of the patient population ranged from 20 to 90 years, mean (SD) 55.3 ± 13.0 years and did not vary across time. Based on medical insurance, the population was underserved: 75% had public-type (Medicare/Medicaid), 15% no insurance, and only 3.1% had private insurance. Health insurance coverage and participant weight significantly changed over time. The percentage of patients whose insurance type was not documented—7.2%—increased over time, and the diabetes patient population’s mean weight increased by approximately 10 pounds in the four-year study period (data not shown).
buy prescription drugs without prescription

Table 3. Prevalence of Accountability Patterns of Care Over Time. Project DIRECT, 1996-2000

	Year 1	Year 2	Year3	Year 4	Year 1-Year 4	P Value
					Relative Difference,! %	Association = 0	Linear Trend = 0
HbAlС tested %with HbAl с >9.5%	54.5 55.2	76.4 27.8	92.6 57.3	91.7 33.5	68 -39	0.0001 0.008	0.0009 0.29
HbAl с periodicity t 0/year§ 1 /year >l/year	6.9 45.5 47.5	6.5 40.4 53.0	6.2 29.3 64.4	7.0 28.7 64.4	1 -37 36	0.002	0.006
Referral for eye exam	32.5	52.5	50.3	71.5	120	0.0001	0.0001
HDL/LDL cholesterol tested	85.7	74.7	79.3	69.8	-29	0.10	0.02
Diabetic nephropathy Microalbuminuria test Proteinuria test	3.2 66.2	10.0 76.2	2.5 79.4	17.3 76.5	440 16	0.02 0.24	0.02 0.09
Blood pressure control, % Normal BP (<140/90 mm Hg)	49.1	67.3	53.3	46.1	-6	0.07	0.54
Foot skin integrity exam	66.8	87.1	78.8	84.0	26	0.05	0.04
Smoking cessation counseling	23.0	33.7	35.9	44.3	93	0.52	0.13
t Relative difference = [(value4-valuel)/valuelj; t Not DQIP variable; § Periodicity of HbAl с not adjusted for intrapractice/physician correlation because of multilevel categories.

At baseline, BP and weight measurement, HDL/LDL and glucose testing, and diet assessment were at the highest end of documented care (75-100%), whereas microalbuminuria was at the lowest (3.2%) (Tables 3-5). HbAlc analyses, foot skin integrity, and vascular examination were documented in 50-75% of charts, referral for dilated eye exam in about one-third of charts, and foot care counseling in about one-fifth.
generic viagra online

Table 4. Prevalence of Quality Improvement Patterns of Care over Time. Project DIRECT, 1996-2000

	Year 1	Year 2	Year3	Year 4	Year 1-Year 4 Relative Difference,! %	P Value
						Association = 0	Linear Trend = 0
HbAl с categories, %t
<7%	56.5	38.2	23.6	27.8	-51	0.0004	0.001
7-7.9%	4.7	11.2	10.9	15.9	238
8-8.9%	9.4	16.9	17.3	12.7	35
9-9.9%	1.2	7.9	10.9	11.1	93
>10%	28.2	25.8	37.3	32.5	15
Systolic blood pressure,
mmHg, %\
<140	49.4	36.0	50.9	50.8	3	0.60	0.31
140-159	30.6	44.9	34.6	33.3	1
160-179	15.3	13.5	9.1	11.1	-37
180-209	4.7	4.5	4.6	4.8	0
>210	0.0	1.1	0.9	0.0	0
Diastolic blood pressure,
mmHg, %%
<90	77.7	70.8	76.4	80.1	16	0.03	0.04
90-99	8.2	23.6	16.4	14.3	74
100-109	7.1	5.6	5.5	3.2	-55
110-119	4.7	0.0	1.8	1.6	-66
>120	2.4	0.0	0.0	0.0	0
Diet assessment	77.1	87.6	76.3	72.9	-5	0.11	0.30
PA assessment	33.2	52.6	28.4	39.5	19	0.06	0.91
Diabetes education	49.0	91.7	87.6	92.0	88	0.001	0.0001
HBGM assessment	58.6	78.2	87.2	89.4	53	0.03	0.0001
Hypoglycemia assessment	43.3	35.4	32.5	38.1	-12	0.71	0.48
Foot care counseling	17.7	55.2	28.0	50.0	82	0.001	0.10
t Relative difference = [(value4-valuel)/valuel]; % Not DQIP variable; § Periodicity of HbAl с not adjusted for intrapractice/physician correlation because of multilevel categories; PA: physical activity; HBGM: home blood glucose monitoring

The prevalence of most of the accountability patterns of care increased over time, well above the pro ject’s goal of a 10% yearly change, for most of them (Table 3). The increase was statistically significant, however, only for HbAlc, referral for eye exam, and microalbuminuria test. During the study period, providers tended to increase the periodicity of HbAlc testing to more than once a year. The prevalence of controlled diabetes decreased, and only part of the indicators of patient lifestyle and diabetes self-management behavior showed statistically significant increases (Table 4). Documentation of peripheral nerve exam decreased, but monofilament use for neurological examination increased from no use at baseline to 8% (p <0.05) use four years later (Table 5).

The periodicity of HbAlc and blood glucose testing and vascular exam increased over time. The median number of times per year these assessments were performed increased from one to two for HbAlc (annual range, 0-6) and vascular exam (annual range, low 0-5 to high 0-14), and from three to four for blood glucose (annual range, low 0-12 to high 0-35). Referral of diabetes patients or persons at risk for diabetes to DIRECT risk-reduction programs was very poor (less than 5%) during the study period. viagra soft

Table 5. Prevalence of Provider’s Other Assessment Activities and Cares over Time. The Project DIRECT, 1996-2000

	Year 1	Year 2	Year3	Year 4	Year 1-Year 4 Relative Difference,! %	P Value
						Association = 0	Linear Trend = 0
Medication Evaluation
ACE inhibitors	39.8	55.2	66.3	58.1	46	0.01	0.01
Insulin	52.4	54.6	59.4	55.9	7	0.86	0.54
Hypoglycemic oral agent	64.6	71.0	82.7	83.8	30	0.01	0.001
Complications Monitoring
Peripheral nerve exam	32.5	67.3	44.9	27.2	-16	0.0001	0.28
Monofilament usej	0	5.0	0.9	8.0	8	0.03	0.06
Vascular exam	50.9	74.9	81.4	83.0	63	0.0001	0.0001
Depression assessment	13.5	19.6	7.1	13.2	CN	0.26	0.56
t Relative difference = [(value4-valuel)/valuel]; t Not DQIP variable; § Periodicity of HbAl с not adjusted for intrapractice/physician correlation because of multilevel categories; ACE: Angiotensin Converting Enzyme

Glycemic Control over Time

Based on cross-sectional samples of patients in practices that stayed in the program for four years, mean HbAlc did not vary and remained at 9.9% ± 0.7%, 9.0% ± 0.7%, 10.2% ± 0.6%, and 9.2% ± 0.6% at years one through four (p=0.17). cialis soft tablets

Office System Change Variables

Use of reminder systems by providers or practices reflective of office system changes was low initially and increased over time. Sticker use increased from no use at baseline to about 35% in the second year and thereafter (p <0.05). Flow sheet use increased from 17.1% at baseline to 38.2% in the fourth year among the providers who stayed the four-year study period (p O.05).