- Email: [email protected]
The Receipt of Prevention Services by Veterans Using VA versus Non-VA Facilities
27, 690–696 (1998) PM980345
PREVENTIVE MEDICINE ARTICLE NO.
The Receipt of Prevention Services by Veterans Using VA versus Non-VA Facilities Donna J. Rabiner, Ph.D.,*,† Laurence G. Branch, Ph.D.,*,†,1 and Robert J. Sullivan, Jr., M.D., M.P.H.*,† *National Center for Health Promotion, Veterans Administration Medical Center, 508 Fulton Street, Durham, North Carolina 27705; and †Center for the Study of Aging and Human Development, Duke University Medical School, Campus Box 3003, Durham, North Carolina 27710
INTRODUCTION
Objectives. This paper compares the health promotion/disease prevention services received by veterans who reported receiving 90%1 of their care inside Veterans Health Administration (VA) facilities with counterparts who reported receiving 90%1 of their care outside VA facilities. Results are compared with the U.S. Healthy People 2000 goals. Methods. Random samples were drawn of 300 men and 150 women visiting primary care clinics in six VA facilities. A 66% adjusted response rate was achieved after two mailings (n 5 1,703). For this analysis, those veterans who reported receiving 90%1 of their care inside VA facilities (n 5 909) were compared with veterans who reported receiving 90%1 of their care outside VA facilities (n 5 185). Results. Of the 13 health promotion/disease prevention services, 6 were significantly influenced by source of care. Five of the significant differences reflected statistically higher prevalence rates for those receiving 90%1 of their care inside the VA (mammograms and counseling for alcohol, nutrition, exercise, and seatbelt use). One reflected a higher prevalence rate for those receiving 90%1 of their care outside the VA system (tetanus boosters). Conclusions. Veterans receiving 90%1 of their care in VA facilities obtained more preventive services than counterparts using non-VA providers. Assessment and counseling services need to be targeted to more veterans to comply more fully with U.S. Preventive Services Task Force recommendations and Healthy People 2000 objectives. q1998 American Health Foundation and Academic Press Key Words: health promotion; disease prevention; veterans; Healthy People 2000.
1 To whom reprint requests should be addressed. Fax: (919) 4165879. E-mail: [email protected].
The U.S. Preventive Services Task Force (USPSTF) recently summarized the current research on the effectiveness and efficiency of a variety of individual health promotion and disease prevention services. The USPSTF based its recommendations on the quality of the available evidence [1] and has challenged U.S. health care providers to respond. The Veterans Health Administration (VA), one of the largest health care systems in the United States, recognizes the importance of providing comprehensive, coordinated, cost-effective preventive, palliative, and curative care services for its U.S. veteran patient population [2]. Given the size and complexity of the VA system, exemplified by the fact that it processed 926,000 inpatient admissions and over 27,500,000 outpatient visits in fiscal year 1995 [3], finding ways to promote healthy behaviors and practices for its male and female veterans has become a major management, clinical, and administrative policy objective [2,4]. The VA’s National Center for Health Promotion and Disease Prevention (NCHP) has applied the evidencebased USPSTF recommendations as the standards of care for the asymptomatic, average-risk veteran receiving primary care in VA facilities [5]. The 13 recommended services comprise the core health promotion and disease prevention services that all average-risk veterans can expect to receive from their primary care provider. Primary prevention services for the VA include: (1) hypertension detection, (2) hyperlipidemia detection, (3) influenza immunization, (4) pneumococcal vaccination, and (5) tetanus and diphtheria immunization. Secondary prevention services include early detection and treatment of: (6) cervical cancer, (7) breast cancer, and (8) colorectal cancer. Assessment and counseling, if appropriate, are recommended with regard to: (9) tobacco
690
0091-7435/98 $25.00 Copyright q 1998 by American Health Foundation and Academic Press All rights of reproduction in any form reserved.
691
SERVICES AT VA VERSUS NON-VA FACILITIES
use, (10) problem drinking, (11) weight control and nutrition, (12) fitness and exercise, and (13) seatbelt use and accident avoidance. While it would be ideal from the perspective of personal health if compliance with each VA health promotion and disease prevention guideline was 100%, in most cases this is not practical. Accordingly, measurable goals are modeled after those described in Healthy People 2000, [6] a national initiative to improve the health of all Americans through prevention. Healthy People 2000, which is driven by 300 specific national health promotion and disease prevention objectives targeted for achievement by the Year 2000, defines three goals to help the United States reach its full potential: to increase the average span of healthy life, to reduce disparities in health among U.S. citizens, and to provide access to health maintenance activities for all Americans [7,8]. Although it is recognized that Healthy People 2000 goals apply to the entire population while the VA rates to be reported in this study apply only to veterans who visited primary care clinics during the previous year, the Healthy People 2000 objectives have been adapted as the targeted Year 2000 goals within the VA. Therefore, the Healthy People 2000 objectives were selected for comparative purposes to track the VA’s progress in delivering health promotion/disease prevention services to the male and female veterans who obtained medical care at VA primary care clinics. Although prior research has examined a broad range of health services research outcomes among U.S. veterans, including average lengths of stay among veterans using VA versus other systems of care, [9] VA versus Medicare utilization of hospital services among older male veterans in New England and New York, [10] and veterans’ and nonveterans’ use of ambulatory and short-term care in VA facilities, [11] prior work has not investigated the extent to which health promotion/ disease prevention services are received by veterans relying primarily on VA versus non-VA health care providers. Therefore, the purpose of this research is to examine whether and to what extent veterans who receive the majority of their care inside the VA receive a fuller array of health promotion/disease prevention services than their veteran counterparts who rely primarily on non-VA health care providers for their health care. The following three questions have been investigated: (1) Do veterans who report 90% or more of their care from the VA report more of the USPSTF’s recommended health promotion/disease prevention services than veterans who report 90% or more of their care outside of VA facilities? (2) Are veterans who report 90% or more of their care from the VA more likely to be in compliance with the health promotion/disease prevention service
goals outlined in the Healthy People 2000 report? (3) What is the relationship between the location of the receipt of veterans’ general medical care and the location at which the 13 health promotion/disease prevention services were reportedly received? METHODS
Sample Six VA facilities located in Durham, North Carolina; Fresno, Long Beach, and Loma Linda, California; Manhattan, New York; and Seattle, Washington volunteered to test the first implementation of a new survey process and data collection procedure. These sites are located in large-, middle-, and small-sized communities, dispersed geographically in northeast, southeast, and west coast locations. Random samples of 300 men and 150 women were drawn from the list of individuals receiving primary care in any of the following VA outpatient facilities during the previous year: general internal medicine clinics, women’s health clinics, geriatrics clinics, and primary care medicine clinics. Procedures The outpatient encounter file provided the sampling frame used for each VA facility. A complete listing of veterans receiving primary care at each of the six VA facilities between March 1, 1995, and February 29, 1996, was compiled. Random samples of 300 men and 150 females were then drawn representing unique users (not weighted by the amount of use). Identification numbers were assigned to each person sampled and used in lieu of names to preserve the anonymity of potential study respondents. The initial mailing of the Veterans Health Survey was sent to the residential addresses of the selected veterans in the spring of 1996. Nonrespondents received a second mailing approximately 4 weeks later. The deadline for the receipt of completed questionnaires was mid-September 1996. A 66% adjusted response rate was achieved after two mailings (1,703 respondents in total from 2,700 initial sample members minus 131 sample members deceased or unable to be located). Although the same number of men (300) and women (150) were sampled at each VA facility, the total number of men and women actually receiving VA services (and thus, eligible to be included in the sampling frame) varied by VA setting. Therefore, the calculated probability of selection into the sample was different for men and women and different at each VA facility. Consequently, weighting was necessary to obtain proper estimates of rates for the entire VA population.
692
RABINER, BRANCH, AND SULLIVAN
Questionnaire The NCHP Veterans Health Survey questionnaire was designed to obtain information directly from the men and women who visit VA primary care clinics about whether each of the 13 recommended services was received, when it was received, and where it was received (within VA facilities or non-VA sources). The survey approach avoids a potential undercounting bias of health promotion services that would result if VA medical records were used as the sole source of information, because services received outside VA facilities would be omitted. This 52-item multiple-choice questionnaire modeled the structure of the Department of Health and Human Service National Health Interview and its other national surveys and asked a series of questions about each recommended prevention service. The first question in each series asked whether the veteran had ever received a specific procedure or test (yes/no). Those answering “yes” to this item were asked how long it had been since the specific procedure/test was last performed. The interval of time assessed varied by survey question, but it always included the time period specifically recommended by the USPSTF for each health promotion/disease prevention service item. For example, although the USPSTF recommends the receipt of a tetanus booster at least once every 10 years, the actual questionnaire responses to the question “when was the last time (you received a tetanus booster)?” were as follows: 5 or fewer years ago, 6–10 years ago, 11–15 years ago, 16–20 years ago, 21 or more years ago, and don’t recall. Survey response categories were combined in the analysis in order to be directly comparable to USPSTF recommendations. Finally, the respondents who reported receiving the specific procedure/test were asked to indicate where the procedure had been performed (at a VA facility or elsewhere). The estimated average time to complete the questionnaire was 10 minutes. Data Processing Data from the six sites were entered, 100% verified, and cleaned. Test–retest k reliability statistics were calculated for the 52 items on the questionnaire using data from 32 veterans who happened to complete the survey twice, presumably because their first response crossed in the mail with a second request to which they also responded. Thirty-seven percent of the 52 survey questions had k scores greater than 0.80, another 33% were greater than 0.60, and another 19% were greater than 0.40, which exceeds the customary threshold for acceptable reliability [12]. Of the remaining 6 questions, 5 had dispersion characteristics that rendered the k statistic inadequate as an indicator of reliability.
Data Analysis For this analysis, those veterans who reported receiving 90% or more of their care in VA facilities (Mainly in VA, n 5 909) were compared with those veterans who reported receiving 90% or more of their care in non-VA facilities (Mainly in non-VA, n 5 185) in order to examine the impact of source of care on the likelihood of obtaining the 13 prevention services recommended by the USPSTF. Two sets of analyses were conducted. In the first set, prevalence rates for those receiving care Mainly in VA and Mainly in non-VA facilities were calculated for each of the 13 prevention services using the SAS program for microcomputers [13]. Prevalence rates were produced using data stratified by the reported source of their care to examine differences in the responses to specific study questions. x2 statistics were used to assess the statistical significance of prevalence rates for the receipt of primary health promotion/ disease prevention services by usual source of care (i.e., Mainly in VA versus Mainly in non-VA). A significance level of P , 0.05 was selected for all x2 statistics. In the second set of analyses, prevalence rates were generated to examine where care was received for each of the 13 health promotion/disease prevention services, stratified by usual source of care. The following locations of care were examined: recommended service received within a VA facility, recommended service received outside a VA facility, recommended service received both inside and outside a VA facility, and recommended service not received (“no care”). RESULTS
The prevalence rates for the 13 health promotion/ disease prevention services are shown in Table 1 for those veterans receiving care Mainly in VA and Mainly in non-VA facilities. Among the primary prevention services, those veterans reporting receiving care Mainly in non-VA facilities were 12.6% more likely than those receiving care Mainly in VA facilities to have received a tetanus booster within the past decade. Among secondary prevention services, females ages 50–69 were 27.0% more likely to have received a mammogram within the past 2 years if they had obtained care Mainly in VA facilities. In regard to assessment and counseling, veterans who reported care Mainly in VA facilities were significantly more likely to be current smokers and significantly less likely to report “almost always” wearing seatbelts than counterparts receiving care Mainly in non-VA facilities. Notwithstanding these differences in risk status, veterans receiving care Mainly in VA facilities were significantly more likely to receive assessment and counseling services (as shown in Table 1). The prevalence rates for four assessment and counseling services favored the VA system (i.e., alcohol counseling,
693
SERVICES AT VA VERSUS NON-VA FACILITIES
TABLE 1 Prevalence Rates for the Receipt of Health Promotion/Disease Prevention Services by Source of Care Prevalence rates Health promotion/disease prevention service Primary prevention 1 Received blood pressure check within the past 2 years 2a Received cholesterol check in the past 5 years (males ages 35–65) 2b Received cholesterol check in the past 5 years (females ages 45–65) 3 Received influenza vaccine this year (ages 651) 4 Received pneumococcal vaccine at least once (ages 651) 5 Received tetanus booster at least once in the past decade Secondary prevention 6 Females receiving Pap smear test in the past 3 years (under age 65) 7 Females receiving mammograms in the past 2 years (ages 50–69) 8 Received fecal occult blood test this year (ages 501) Assessment and counseling 9a Current smokers 9b Tobacco users offered counseling 10 Received alcohol counseling this year 11 Received nutrition counseling this year 12 Received exercise counseling this year 13a Almost always using seatbelts 13b Received seatbelt counseling this year a b
Year 2000 goals
Veterans reporting 90%1 care in the VA (n 5 909)
Veterans reporting 90%1 care outside the VA (n 5 185)
P value
90%
96.0%
94.0%
0.225
75%
81.0%
84.8%
0.503
75%
85.6%
71.4%
0.317
60% 60%
75.2% 49.5%
68.8% 42.6%
0.286 0.315
62%
58.0%
70.6%
0.002
85%
93.4%
95.8%
0.519
60%
93.7%
66.7%
0.020
50%
30.2%
36.3%
0.239
15%a 75%b 75%b 75%b 50%b 85% 50%b
27.9% 84.6% 15.3% 37.3% 50.1% 83.7% 6.0%
20.5% 81.6% 3.2% 18.9% 28.6% 93.5% 0.5%
0.038 0.628 0.001 0.001 0.001 0.001 0.002
Reduction goal. Adapted from percentage of physicians offering the service to the percentage receiving the service.
nutrition counseling, exercise counseling, and seatbelt counseling). The two differences observed in identifying the at-risk subgroups (i.e., seatbelt nonusers and tobacco users) were a reflection of the baseline differences in the two subgroups studied rather than the services being offered by the respective health care settings. When the prevalence rates for veterans receiving the majority of their care inside or outside the VA were compared with Year 2000 goals (column 1 in Table 1), both groups of veterans were found to exceed the Year 2000 goals for the following primary prevention services: blood pressure checks within the past 2 years, cholesterol checks for males ages 35–65, and influenza vaccines for those ages 651 within the past year. Neither veteran group met the Year 2000 objectives for the receipt of pneumoccocal vaccines at least once for those ages 651. Finally, while veterans receiving care Mainly in non-VA facilities exceeded the Year 2000 objectives for the receipt of a tetanus booster within the past decade, those veterans reporting receiving care Mainly in VA facilities did not meet this objective.
Both groups of veterans exceeded the Year 2000 objectives for the following two secondary prevention services: Pap smear tests within the past 3 years for females under age 65 and mammograms within the past 2 years for females ages 50–69. Neither veteran subgroup met the Year 2000 objectives for the receipt of fecal occult blood tests this year for veterans ages 50 and over. Table 2 presents cross-tabulations of the 13 health promotion/disease prevention services by where the service was received (inside VA, outside VA, both inside and outside VA, and no care), stratified by usual source of care. Among the subgroup reporting receiving care Mainly in VA facilities, most primary and secondary prevention services were obtained primarily inside VA primary care clinics, indicating that most veterans using the VA system for their overall health care needs also are obtaining their preventive services at VA facilities. Interestingly, sizable proportions of those receiving
694
RABINER, BRANCH, AND SULLIVAN
TABLE 2 Prevalence Rates for the Receipt of Health Promotion/Disease Prevention Services by Site of Care 90%1 care in VA Health promotion/disease prevention service Primary prevention 1 Received blood pressure check in the past 2 years 2a Received cholesterol check in the past 5 years (males 35–65) 2b Received cholesterol check in the past 5 years (females 45–65) 3 Received influenza vaccine this year (ages 651) 4 Received pneumoccocal vaccine at least once (ages 651) 5 Received tetanus booster at least once in the past decade Secondary prevention 6 Females received Pap smear in the past 3 years (under age 65) 7 Females received mammograms in the past 2 years (ages 50–69) 8 Received fecal occult blood test this year (ages 501) Assessment and counseling 9 Tobacco users offered counseling this year 10 Received alcohol counseling this year 11 Received nutrition counseling this year 12 Received exercise counseling this year 13 Received seatbelt counseling this year
90%1 care outside VA
(n)
VA
Outside VA
Both
No care
(n)
VA
(889)
76.8
0.8
20.1
2.3
(178)
16.9
(250)
68.8
7.2
7.2
16.8
(59)
(96)
71.9
5.2
9.4
13.5
(402)
63.7
11.4
—
(381)
34.7
15.2
(766)
23.6
(224)
Both
No care
34.8
45.5
2.8
8.5
59.3
17.0
15.2
(11)
14.2
28.6
28.6
28.6
24.9
(60)
18.3
51.7
—
30.0
—
50.1
(59)
13.6
28.8
—
57.6
33.4
—
43.0
(112)
6.4
60.5
—
33.1
79.0
6.3
8.0
6.7
(46)
23.9
50.0
21.7
4.4
(79)
74.7
12.7
6.3
6.3
(6)
33.3
16.7
16.7
33.3
(613)
27.6
2.1
1.6
68.7
(86)
5.8
29.1
2.3
62.8
(251)
68.5
3.6
12.8
15.1
(38)
18.4
47.4
15.8
18.4
(735)
16.5
0.5
1.5
81.5
(150)
0.0
1.3
2.7
96.0
(858)
35.2
0.6
3.7
60.5
(173)
4.0
13.3
2.9
79.8
(812)
49.5
1.4
4.6
44.5
(160)
6.3
19.4
6.3
68.0
(845)
5.0
0.7
0.4
93.9
(173)
0.0
0.6
0.0
99.4
care Mainly in non-VA facilities reported receiving several primary and secondary prevention services both inside and outside of VA facilities. This contrasts directly with the proportions reporting receiving services both inside and outside the VA system for the subgroup reporting care Mainly in VA facilities. (Only 20% of those with care Mainly in VA facilities reported receiving blood pressure checks within the past 2 years both inside and outside the VA system, only 9% of females ages 45–65 received cholesterol checks within the past 5 years both inside and outside the VA system, and only 8% of females under age 65 reported receiving Pap smears within the past 3 years both inside and outside the VA system.) A very small proportion of the veterans receiving care Mainly in non-VA facilities reported receiving assessment and counseling services within the past year (inside, outside, or both inside and outside of VA facilities). While only 18.4% of current tobacco users reported receiving no counseling services within the past year, 96%
Outside VA
of all veterans sampled reported receiving no alcohol counseling services, almost 80% reported receiving no nutrition counseling services, 68% reported receiving no exercise counseling services, and over 99% reported receiving no seatbelt counseling services within the past year. While sizable proportions of the subgroup reporting receiving their care Mainly in VA facilities also reported receiving no counseling services within the past year, these rates were much less pronounced when compared with veteran counterparts receiving care Mainly in nonVA facilities. Surprisingly, a sizable proportion of the veteran subsample reporting receiving care Mainly in VA facilities obtained pneumococcal vaccines (for those age 651) and tetanus boosters outside the VA system (15.2 and 33.4%, respectively). Similarly, a sizable proportion of the female veteran subsample reporting receiving care Mainly in non-VA settings obtained Pap smears within
695
SERVICES AT VA VERSUS NON-VA FACILITIES
the past 3 years (for those under age 65) and mammograms within the past 2 years (for those ages 50–69) inside the VA (23.9 and 33.3%, respectively). Finally, a large proportion of the male and female smoking subsamples with care Mainly in non-VA settings obtained tobacco counseling services inside the VA (18%). Therefore, at least for these five preventive medical services, many of those relying primarily on the VA system revert to non-VA providers to obtain their periodic immunizations, while many of those depending primarily on nonVA providers turn to VA providers to obtain their women’s preventive health and tobacco counseling services. DISCUSSION
This paper examined the extent to which the veterans health system has exceeded the non-VA system in providing its veteran population with the 13 health promotion/disease prevention services recommended by the USPSTF. As one of the largest health care systems in the United States, the VA is committed to providing quality, cost-effective primary, secondary, and tertiary prevention services to its male and female veterans. Learning to what extent the VA system is complying with both USPSTF recommendations and the Year 2000 objectives is important since the VA system processes almost 1 million inpatient and over 27 million outpatient admissions in the United States each year. Overall, veterans reporting receiving care Mainly in VA facilities did as well or better than counterparts receiving care Mainly in non-VA facilities. When the eight significant differences between these two groups were examined, four reflected statistically higher prevalence rates achieved among those receiving care primarily inside the VA in assessment and counseling services and the fifth was the receipt of mammograms every 2 years for females ages 50–69. One—the receipt of tetanus boosters within the past decade—reflected significantly higher prevalence rates among non-VA users. The remaining two significant differences [use of tobacco (a reduction target) and “almost always” using seat belts] indicated that non-VA users did better in these health behaviors relative to counterparts receiving care primarily inside the VA system, which reinforces the contention that VA patients are often sicker than their non-VA counterparts. Overall, although the VA system did relatively better than the non-VA system in promoting most health promotion/disease prevention services, additional work is needed to comply more fully with the USPSTF recommendations for all assessment and counseling services. Comparisons within each of the two veteran subsamples indicated not surprisingly that those receiving care primarily inside the VA system were more likely to receive the health promotion/disease prevention services inside the VA system, while those receiving care
primarily outside the VA system were more likely to rely on non-VA providers. Of greater concern was the fact that those receiving care primarily outside the VA system generally had higher rates of no care for the health promotion/disease prevention services targeted by the USPSTF and specified in the Healthy People 2000 report. This finding may reflect a fragmentation or lack of primary care providers among those veterans who reported receiving 90%1 of their care outside the VA. Unfortunately, we have no way of knowing which veterans receiving 90%1 of their care outside may be members of managed care organizations or have established a relationship with a primary care provider. In summary, based on the cumulative evidence presented in this paper, the VA system appears to be successful in providing veterans with each of the 13 recommended health promotion/disease prevention services. Although there remain several areas where additional work is needed, these data indicate that preventive medical services received by both male and female veterans inside the VA system currently exceed the U.S. Year 2000 Goals in the areas of hypertension detection, hyperlipidemia detection, and influenza immunization. Preventive medical services for female veterans receiving 90%1 of their care inside the VA system also exceed the U.S. Year 2000 Goals in the areas of cervical cancer and breast cancer detection. While assessment and counseling services currently are more likely to be obtained by male and female veterans receiving at least 90% of their care inside the VA system, more attention in this area is still warranted since all five screening and counseling services provided to U.S. veterans have not been received at the levels recommended in the Healthy People 2000 report. The NCHP currently is implementing the Veterans Health Survey in all VA facilities in Fiscal Year 1997. These data will provide a baseline for assessing future improvements in the provision of health promotion and disease prevention services. The larger study will further our understanding of the rates of preventive medical services received by veterans using VA and non-VA facilities and help to target efforts to those most likely to receive substantial benefits. REFERENCES 1. U.S. Preventive Services Task Force Guide to clinical preventive services. 2nd ed. Baltimore: William & Williams Publ., 1996. 2. Kizer, KW. Journey of change. Washington: Veterans Administration, 1997. 3. Department of Veterans Affairs. Summary of medical programs: fiscal year 1995: cumulative October 1, 1994 through September 30, 1995. Internet address: http://www.va.gov/sumedprf/fy95/ sumfy95.htm. 4. Kizer, KW. Prescription for change: the guiding principles and strategic objectives underlying the transformation of the veterans healthcare system. Washington: Veterans Administration, 1996.
696
RABINER, BRANCH, AND SULLIVAN
5. VA handbook 1101.8. Health promotion and disease prevention (RCN-10-06666). Washington: U.S. Govt. Printing Office, 1996. 6. U.S. Department of Health and Human Services, Public Health Service. Healthy People 2000: national health promotion and disease prevention objectives. Washington: U.S. Govt. Printing Office, 1991. [DHHS Publication No. (PHS) 91-500212] 7. U.S. Department of Health and Human Services, Public Health Service. Healthy People 2000 update: achieving the nation’s health objectives. Washington: Office of Disease Prevention and Health Promotion, 1993 Feb/Mar. 8. Sullivan LW. Sounding board. Healthy People 2000. N Engl J Med 1990;23(15):1065.
9. Wolinsky D, Coe MM, Mosely, RR. Length of stay in the VA: longterm care in short-term hospitals. Med Care 1987;25(3):250. 10. Fleming C, Fisher ES, Chang C, Bubolz TA, et al. Studying outcomes and hospital utilization in the elderly. Med Care 1992; 30(5):377. 11. Wolinsky FD, Coe RM, Mosely RR, et al. Veterans and nonveterans’ use of health services: a comparative analysis. Med Care 1985;23(12):1358. 12. Fleiss JL. Statistical methods for rates and proportions. 2nd ed. New York: Wiley, 1981. 13. SAS Institute. SAS/STAT user’s guide. Vol. 1 and 2. Version 6. Cary (NC): SAS Institute, 1990.
PREVENTIVE MEDICINE ARTICLE NO.
The Receipt of Prevention Services by Veterans Using VA versus Non-VA Facilities Donna J. Rabiner, Ph.D.,*,† Laurence G. Branch, Ph.D.,*,†,1 and Robert J. Sullivan, Jr., M.D., M.P.H.*,† *National Center for Health Promotion, Veterans Administration Medical Center, 508 Fulton Street, Durham, North Carolina 27705; and †Center for the Study of Aging and Human Development, Duke University Medical School, Campus Box 3003, Durham, North Carolina 27710
INTRODUCTION
Objectives. This paper compares the health promotion/disease prevention services received by veterans who reported receiving 90%1 of their care inside Veterans Health Administration (VA) facilities with counterparts who reported receiving 90%1 of their care outside VA facilities. Results are compared with the U.S. Healthy People 2000 goals. Methods. Random samples were drawn of 300 men and 150 women visiting primary care clinics in six VA facilities. A 66% adjusted response rate was achieved after two mailings (n 5 1,703). For this analysis, those veterans who reported receiving 90%1 of their care inside VA facilities (n 5 909) were compared with veterans who reported receiving 90%1 of their care outside VA facilities (n 5 185). Results. Of the 13 health promotion/disease prevention services, 6 were significantly influenced by source of care. Five of the significant differences reflected statistically higher prevalence rates for those receiving 90%1 of their care inside the VA (mammograms and counseling for alcohol, nutrition, exercise, and seatbelt use). One reflected a higher prevalence rate for those receiving 90%1 of their care outside the VA system (tetanus boosters). Conclusions. Veterans receiving 90%1 of their care in VA facilities obtained more preventive services than counterparts using non-VA providers. Assessment and counseling services need to be targeted to more veterans to comply more fully with U.S. Preventive Services Task Force recommendations and Healthy People 2000 objectives. q1998 American Health Foundation and Academic Press Key Words: health promotion; disease prevention; veterans; Healthy People 2000.
1 To whom reprint requests should be addressed. Fax: (919) 4165879. E-mail: [email protected].
The U.S. Preventive Services Task Force (USPSTF) recently summarized the current research on the effectiveness and efficiency of a variety of individual health promotion and disease prevention services. The USPSTF based its recommendations on the quality of the available evidence [1] and has challenged U.S. health care providers to respond. The Veterans Health Administration (VA), one of the largest health care systems in the United States, recognizes the importance of providing comprehensive, coordinated, cost-effective preventive, palliative, and curative care services for its U.S. veteran patient population [2]. Given the size and complexity of the VA system, exemplified by the fact that it processed 926,000 inpatient admissions and over 27,500,000 outpatient visits in fiscal year 1995 [3], finding ways to promote healthy behaviors and practices for its male and female veterans has become a major management, clinical, and administrative policy objective [2,4]. The VA’s National Center for Health Promotion and Disease Prevention (NCHP) has applied the evidencebased USPSTF recommendations as the standards of care for the asymptomatic, average-risk veteran receiving primary care in VA facilities [5]. The 13 recommended services comprise the core health promotion and disease prevention services that all average-risk veterans can expect to receive from their primary care provider. Primary prevention services for the VA include: (1) hypertension detection, (2) hyperlipidemia detection, (3) influenza immunization, (4) pneumococcal vaccination, and (5) tetanus and diphtheria immunization. Secondary prevention services include early detection and treatment of: (6) cervical cancer, (7) breast cancer, and (8) colorectal cancer. Assessment and counseling, if appropriate, are recommended with regard to: (9) tobacco
690
0091-7435/98 $25.00 Copyright q 1998 by American Health Foundation and Academic Press All rights of reproduction in any form reserved.
691
SERVICES AT VA VERSUS NON-VA FACILITIES
use, (10) problem drinking, (11) weight control and nutrition, (12) fitness and exercise, and (13) seatbelt use and accident avoidance. While it would be ideal from the perspective of personal health if compliance with each VA health promotion and disease prevention guideline was 100%, in most cases this is not practical. Accordingly, measurable goals are modeled after those described in Healthy People 2000, [6] a national initiative to improve the health of all Americans through prevention. Healthy People 2000, which is driven by 300 specific national health promotion and disease prevention objectives targeted for achievement by the Year 2000, defines three goals to help the United States reach its full potential: to increase the average span of healthy life, to reduce disparities in health among U.S. citizens, and to provide access to health maintenance activities for all Americans [7,8]. Although it is recognized that Healthy People 2000 goals apply to the entire population while the VA rates to be reported in this study apply only to veterans who visited primary care clinics during the previous year, the Healthy People 2000 objectives have been adapted as the targeted Year 2000 goals within the VA. Therefore, the Healthy People 2000 objectives were selected for comparative purposes to track the VA’s progress in delivering health promotion/disease prevention services to the male and female veterans who obtained medical care at VA primary care clinics. Although prior research has examined a broad range of health services research outcomes among U.S. veterans, including average lengths of stay among veterans using VA versus other systems of care, [9] VA versus Medicare utilization of hospital services among older male veterans in New England and New York, [10] and veterans’ and nonveterans’ use of ambulatory and short-term care in VA facilities, [11] prior work has not investigated the extent to which health promotion/ disease prevention services are received by veterans relying primarily on VA versus non-VA health care providers. Therefore, the purpose of this research is to examine whether and to what extent veterans who receive the majority of their care inside the VA receive a fuller array of health promotion/disease prevention services than their veteran counterparts who rely primarily on non-VA health care providers for their health care. The following three questions have been investigated: (1) Do veterans who report 90% or more of their care from the VA report more of the USPSTF’s recommended health promotion/disease prevention services than veterans who report 90% or more of their care outside of VA facilities? (2) Are veterans who report 90% or more of their care from the VA more likely to be in compliance with the health promotion/disease prevention service
goals outlined in the Healthy People 2000 report? (3) What is the relationship between the location of the receipt of veterans’ general medical care and the location at which the 13 health promotion/disease prevention services were reportedly received? METHODS
Sample Six VA facilities located in Durham, North Carolina; Fresno, Long Beach, and Loma Linda, California; Manhattan, New York; and Seattle, Washington volunteered to test the first implementation of a new survey process and data collection procedure. These sites are located in large-, middle-, and small-sized communities, dispersed geographically in northeast, southeast, and west coast locations. Random samples of 300 men and 150 women were drawn from the list of individuals receiving primary care in any of the following VA outpatient facilities during the previous year: general internal medicine clinics, women’s health clinics, geriatrics clinics, and primary care medicine clinics. Procedures The outpatient encounter file provided the sampling frame used for each VA facility. A complete listing of veterans receiving primary care at each of the six VA facilities between March 1, 1995, and February 29, 1996, was compiled. Random samples of 300 men and 150 females were then drawn representing unique users (not weighted by the amount of use). Identification numbers were assigned to each person sampled and used in lieu of names to preserve the anonymity of potential study respondents. The initial mailing of the Veterans Health Survey was sent to the residential addresses of the selected veterans in the spring of 1996. Nonrespondents received a second mailing approximately 4 weeks later. The deadline for the receipt of completed questionnaires was mid-September 1996. A 66% adjusted response rate was achieved after two mailings (1,703 respondents in total from 2,700 initial sample members minus 131 sample members deceased or unable to be located). Although the same number of men (300) and women (150) were sampled at each VA facility, the total number of men and women actually receiving VA services (and thus, eligible to be included in the sampling frame) varied by VA setting. Therefore, the calculated probability of selection into the sample was different for men and women and different at each VA facility. Consequently, weighting was necessary to obtain proper estimates of rates for the entire VA population.
692
RABINER, BRANCH, AND SULLIVAN
Questionnaire The NCHP Veterans Health Survey questionnaire was designed to obtain information directly from the men and women who visit VA primary care clinics about whether each of the 13 recommended services was received, when it was received, and where it was received (within VA facilities or non-VA sources). The survey approach avoids a potential undercounting bias of health promotion services that would result if VA medical records were used as the sole source of information, because services received outside VA facilities would be omitted. This 52-item multiple-choice questionnaire modeled the structure of the Department of Health and Human Service National Health Interview and its other national surveys and asked a series of questions about each recommended prevention service. The first question in each series asked whether the veteran had ever received a specific procedure or test (yes/no). Those answering “yes” to this item were asked how long it had been since the specific procedure/test was last performed. The interval of time assessed varied by survey question, but it always included the time period specifically recommended by the USPSTF for each health promotion/disease prevention service item. For example, although the USPSTF recommends the receipt of a tetanus booster at least once every 10 years, the actual questionnaire responses to the question “when was the last time (you received a tetanus booster)?” were as follows: 5 or fewer years ago, 6–10 years ago, 11–15 years ago, 16–20 years ago, 21 or more years ago, and don’t recall. Survey response categories were combined in the analysis in order to be directly comparable to USPSTF recommendations. Finally, the respondents who reported receiving the specific procedure/test were asked to indicate where the procedure had been performed (at a VA facility or elsewhere). The estimated average time to complete the questionnaire was 10 minutes. Data Processing Data from the six sites were entered, 100% verified, and cleaned. Test–retest k reliability statistics were calculated for the 52 items on the questionnaire using data from 32 veterans who happened to complete the survey twice, presumably because their first response crossed in the mail with a second request to which they also responded. Thirty-seven percent of the 52 survey questions had k scores greater than 0.80, another 33% were greater than 0.60, and another 19% were greater than 0.40, which exceeds the customary threshold for acceptable reliability [12]. Of the remaining 6 questions, 5 had dispersion characteristics that rendered the k statistic inadequate as an indicator of reliability.
Data Analysis For this analysis, those veterans who reported receiving 90% or more of their care in VA facilities (Mainly in VA, n 5 909) were compared with those veterans who reported receiving 90% or more of their care in non-VA facilities (Mainly in non-VA, n 5 185) in order to examine the impact of source of care on the likelihood of obtaining the 13 prevention services recommended by the USPSTF. Two sets of analyses were conducted. In the first set, prevalence rates for those receiving care Mainly in VA and Mainly in non-VA facilities were calculated for each of the 13 prevention services using the SAS program for microcomputers [13]. Prevalence rates were produced using data stratified by the reported source of their care to examine differences in the responses to specific study questions. x2 statistics were used to assess the statistical significance of prevalence rates for the receipt of primary health promotion/ disease prevention services by usual source of care (i.e., Mainly in VA versus Mainly in non-VA). A significance level of P , 0.05 was selected for all x2 statistics. In the second set of analyses, prevalence rates were generated to examine where care was received for each of the 13 health promotion/disease prevention services, stratified by usual source of care. The following locations of care were examined: recommended service received within a VA facility, recommended service received outside a VA facility, recommended service received both inside and outside a VA facility, and recommended service not received (“no care”). RESULTS
The prevalence rates for the 13 health promotion/ disease prevention services are shown in Table 1 for those veterans receiving care Mainly in VA and Mainly in non-VA facilities. Among the primary prevention services, those veterans reporting receiving care Mainly in non-VA facilities were 12.6% more likely than those receiving care Mainly in VA facilities to have received a tetanus booster within the past decade. Among secondary prevention services, females ages 50–69 were 27.0% more likely to have received a mammogram within the past 2 years if they had obtained care Mainly in VA facilities. In regard to assessment and counseling, veterans who reported care Mainly in VA facilities were significantly more likely to be current smokers and significantly less likely to report “almost always” wearing seatbelts than counterparts receiving care Mainly in non-VA facilities. Notwithstanding these differences in risk status, veterans receiving care Mainly in VA facilities were significantly more likely to receive assessment and counseling services (as shown in Table 1). The prevalence rates for four assessment and counseling services favored the VA system (i.e., alcohol counseling,
693
SERVICES AT VA VERSUS NON-VA FACILITIES
TABLE 1 Prevalence Rates for the Receipt of Health Promotion/Disease Prevention Services by Source of Care Prevalence rates Health promotion/disease prevention service Primary prevention 1 Received blood pressure check within the past 2 years 2a Received cholesterol check in the past 5 years (males ages 35–65) 2b Received cholesterol check in the past 5 years (females ages 45–65) 3 Received influenza vaccine this year (ages 651) 4 Received pneumococcal vaccine at least once (ages 651) 5 Received tetanus booster at least once in the past decade Secondary prevention 6 Females receiving Pap smear test in the past 3 years (under age 65) 7 Females receiving mammograms in the past 2 years (ages 50–69) 8 Received fecal occult blood test this year (ages 501) Assessment and counseling 9a Current smokers 9b Tobacco users offered counseling 10 Received alcohol counseling this year 11 Received nutrition counseling this year 12 Received exercise counseling this year 13a Almost always using seatbelts 13b Received seatbelt counseling this year a b
Year 2000 goals
Veterans reporting 90%1 care in the VA (n 5 909)
Veterans reporting 90%1 care outside the VA (n 5 185)
P value
90%
96.0%
94.0%
0.225
75%
81.0%
84.8%
0.503
75%
85.6%
71.4%
0.317
60% 60%
75.2% 49.5%
68.8% 42.6%
0.286 0.315
62%
58.0%
70.6%
0.002
85%
93.4%
95.8%
0.519
60%
93.7%
66.7%
0.020
50%
30.2%
36.3%
0.239
15%a 75%b 75%b 75%b 50%b 85% 50%b
27.9% 84.6% 15.3% 37.3% 50.1% 83.7% 6.0%
20.5% 81.6% 3.2% 18.9% 28.6% 93.5% 0.5%
0.038 0.628 0.001 0.001 0.001 0.001 0.002
Reduction goal. Adapted from percentage of physicians offering the service to the percentage receiving the service.
nutrition counseling, exercise counseling, and seatbelt counseling). The two differences observed in identifying the at-risk subgroups (i.e., seatbelt nonusers and tobacco users) were a reflection of the baseline differences in the two subgroups studied rather than the services being offered by the respective health care settings. When the prevalence rates for veterans receiving the majority of their care inside or outside the VA were compared with Year 2000 goals (column 1 in Table 1), both groups of veterans were found to exceed the Year 2000 goals for the following primary prevention services: blood pressure checks within the past 2 years, cholesterol checks for males ages 35–65, and influenza vaccines for those ages 651 within the past year. Neither veteran group met the Year 2000 objectives for the receipt of pneumoccocal vaccines at least once for those ages 651. Finally, while veterans receiving care Mainly in non-VA facilities exceeded the Year 2000 objectives for the receipt of a tetanus booster within the past decade, those veterans reporting receiving care Mainly in VA facilities did not meet this objective.
Both groups of veterans exceeded the Year 2000 objectives for the following two secondary prevention services: Pap smear tests within the past 3 years for females under age 65 and mammograms within the past 2 years for females ages 50–69. Neither veteran subgroup met the Year 2000 objectives for the receipt of fecal occult blood tests this year for veterans ages 50 and over. Table 2 presents cross-tabulations of the 13 health promotion/disease prevention services by where the service was received (inside VA, outside VA, both inside and outside VA, and no care), stratified by usual source of care. Among the subgroup reporting receiving care Mainly in VA facilities, most primary and secondary prevention services were obtained primarily inside VA primary care clinics, indicating that most veterans using the VA system for their overall health care needs also are obtaining their preventive services at VA facilities. Interestingly, sizable proportions of those receiving
694
RABINER, BRANCH, AND SULLIVAN
TABLE 2 Prevalence Rates for the Receipt of Health Promotion/Disease Prevention Services by Site of Care 90%1 care in VA Health promotion/disease prevention service Primary prevention 1 Received blood pressure check in the past 2 years 2a Received cholesterol check in the past 5 years (males 35–65) 2b Received cholesterol check in the past 5 years (females 45–65) 3 Received influenza vaccine this year (ages 651) 4 Received pneumoccocal vaccine at least once (ages 651) 5 Received tetanus booster at least once in the past decade Secondary prevention 6 Females received Pap smear in the past 3 years (under age 65) 7 Females received mammograms in the past 2 years (ages 50–69) 8 Received fecal occult blood test this year (ages 501) Assessment and counseling 9 Tobacco users offered counseling this year 10 Received alcohol counseling this year 11 Received nutrition counseling this year 12 Received exercise counseling this year 13 Received seatbelt counseling this year
90%1 care outside VA
(n)
VA
Outside VA
Both
No care
(n)
VA
(889)
76.8
0.8
20.1
2.3
(178)
16.9
(250)
68.8
7.2
7.2
16.8
(59)
(96)
71.9
5.2
9.4
13.5
(402)
63.7
11.4
—
(381)
34.7
15.2
(766)
23.6
(224)
Both
No care
34.8
45.5
2.8
8.5
59.3
17.0
15.2
(11)
14.2
28.6
28.6
28.6
24.9
(60)
18.3
51.7
—
30.0
—
50.1
(59)
13.6
28.8
—
57.6
33.4
—
43.0
(112)
6.4
60.5
—
33.1
79.0
6.3
8.0
6.7
(46)
23.9
50.0
21.7
4.4
(79)
74.7
12.7
6.3
6.3
(6)
33.3
16.7
16.7
33.3
(613)
27.6
2.1
1.6
68.7
(86)
5.8
29.1
2.3
62.8
(251)
68.5
3.6
12.8
15.1
(38)
18.4
47.4
15.8
18.4
(735)
16.5
0.5
1.5
81.5
(150)
0.0
1.3
2.7
96.0
(858)
35.2
0.6
3.7
60.5
(173)
4.0
13.3
2.9
79.8
(812)
49.5
1.4
4.6
44.5
(160)
6.3
19.4
6.3
68.0
(845)
5.0
0.7
0.4
93.9
(173)
0.0
0.6
0.0
99.4
care Mainly in non-VA facilities reported receiving several primary and secondary prevention services both inside and outside of VA facilities. This contrasts directly with the proportions reporting receiving services both inside and outside the VA system for the subgroup reporting care Mainly in VA facilities. (Only 20% of those with care Mainly in VA facilities reported receiving blood pressure checks within the past 2 years both inside and outside the VA system, only 9% of females ages 45–65 received cholesterol checks within the past 5 years both inside and outside the VA system, and only 8% of females under age 65 reported receiving Pap smears within the past 3 years both inside and outside the VA system.) A very small proportion of the veterans receiving care Mainly in non-VA facilities reported receiving assessment and counseling services within the past year (inside, outside, or both inside and outside of VA facilities). While only 18.4% of current tobacco users reported receiving no counseling services within the past year, 96%
Outside VA
of all veterans sampled reported receiving no alcohol counseling services, almost 80% reported receiving no nutrition counseling services, 68% reported receiving no exercise counseling services, and over 99% reported receiving no seatbelt counseling services within the past year. While sizable proportions of the subgroup reporting receiving their care Mainly in VA facilities also reported receiving no counseling services within the past year, these rates were much less pronounced when compared with veteran counterparts receiving care Mainly in nonVA facilities. Surprisingly, a sizable proportion of the veteran subsample reporting receiving care Mainly in VA facilities obtained pneumococcal vaccines (for those age 651) and tetanus boosters outside the VA system (15.2 and 33.4%, respectively). Similarly, a sizable proportion of the female veteran subsample reporting receiving care Mainly in non-VA settings obtained Pap smears within
695
SERVICES AT VA VERSUS NON-VA FACILITIES
the past 3 years (for those under age 65) and mammograms within the past 2 years (for those ages 50–69) inside the VA (23.9 and 33.3%, respectively). Finally, a large proportion of the male and female smoking subsamples with care Mainly in non-VA settings obtained tobacco counseling services inside the VA (18%). Therefore, at least for these five preventive medical services, many of those relying primarily on the VA system revert to non-VA providers to obtain their periodic immunizations, while many of those depending primarily on nonVA providers turn to VA providers to obtain their women’s preventive health and tobacco counseling services. DISCUSSION
This paper examined the extent to which the veterans health system has exceeded the non-VA system in providing its veteran population with the 13 health promotion/disease prevention services recommended by the USPSTF. As one of the largest health care systems in the United States, the VA is committed to providing quality, cost-effective primary, secondary, and tertiary prevention services to its male and female veterans. Learning to what extent the VA system is complying with both USPSTF recommendations and the Year 2000 objectives is important since the VA system processes almost 1 million inpatient and over 27 million outpatient admissions in the United States each year. Overall, veterans reporting receiving care Mainly in VA facilities did as well or better than counterparts receiving care Mainly in non-VA facilities. When the eight significant differences between these two groups were examined, four reflected statistically higher prevalence rates achieved among those receiving care primarily inside the VA in assessment and counseling services and the fifth was the receipt of mammograms every 2 years for females ages 50–69. One—the receipt of tetanus boosters within the past decade—reflected significantly higher prevalence rates among non-VA users. The remaining two significant differences [use of tobacco (a reduction target) and “almost always” using seat belts] indicated that non-VA users did better in these health behaviors relative to counterparts receiving care primarily inside the VA system, which reinforces the contention that VA patients are often sicker than their non-VA counterparts. Overall, although the VA system did relatively better than the non-VA system in promoting most health promotion/disease prevention services, additional work is needed to comply more fully with the USPSTF recommendations for all assessment and counseling services. Comparisons within each of the two veteran subsamples indicated not surprisingly that those receiving care primarily inside the VA system were more likely to receive the health promotion/disease prevention services inside the VA system, while those receiving care
primarily outside the VA system were more likely to rely on non-VA providers. Of greater concern was the fact that those receiving care primarily outside the VA system generally had higher rates of no care for the health promotion/disease prevention services targeted by the USPSTF and specified in the Healthy People 2000 report. This finding may reflect a fragmentation or lack of primary care providers among those veterans who reported receiving 90%1 of their care outside the VA. Unfortunately, we have no way of knowing which veterans receiving 90%1 of their care outside may be members of managed care organizations or have established a relationship with a primary care provider. In summary, based on the cumulative evidence presented in this paper, the VA system appears to be successful in providing veterans with each of the 13 recommended health promotion/disease prevention services. Although there remain several areas where additional work is needed, these data indicate that preventive medical services received by both male and female veterans inside the VA system currently exceed the U.S. Year 2000 Goals in the areas of hypertension detection, hyperlipidemia detection, and influenza immunization. Preventive medical services for female veterans receiving 90%1 of their care inside the VA system also exceed the U.S. Year 2000 Goals in the areas of cervical cancer and breast cancer detection. While assessment and counseling services currently are more likely to be obtained by male and female veterans receiving at least 90% of their care inside the VA system, more attention in this area is still warranted since all five screening and counseling services provided to U.S. veterans have not been received at the levels recommended in the Healthy People 2000 report. The NCHP currently is implementing the Veterans Health Survey in all VA facilities in Fiscal Year 1997. These data will provide a baseline for assessing future improvements in the provision of health promotion and disease prevention services. The larger study will further our understanding of the rates of preventive medical services received by veterans using VA and non-VA facilities and help to target efforts to those most likely to receive substantial benefits. REFERENCES 1. U.S. Preventive Services Task Force Guide to clinical preventive services. 2nd ed. Baltimore: William & Williams Publ., 1996. 2. Kizer, KW. Journey of change. Washington: Veterans Administration, 1997. 3. Department of Veterans Affairs. Summary of medical programs: fiscal year 1995: cumulative October 1, 1994 through September 30, 1995. Internet address: http://www.va.gov/sumedprf/fy95/ sumfy95.htm. 4. Kizer, KW. Prescription for change: the guiding principles and strategic objectives underlying the transformation of the veterans healthcare system. Washington: Veterans Administration, 1996.
696
RABINER, BRANCH, AND SULLIVAN
5. VA handbook 1101.8. Health promotion and disease prevention (RCN-10-06666). Washington: U.S. Govt. Printing Office, 1996. 6. U.S. Department of Health and Human Services, Public Health Service. Healthy People 2000: national health promotion and disease prevention objectives. Washington: U.S. Govt. Printing Office, 1991. [DHHS Publication No. (PHS) 91-500212] 7. U.S. Department of Health and Human Services, Public Health Service. Healthy People 2000 update: achieving the nation’s health objectives. Washington: Office of Disease Prevention and Health Promotion, 1993 Feb/Mar. 8. Sullivan LW. Sounding board. Healthy People 2000. N Engl J Med 1990;23(15):1065.
9. Wolinsky D, Coe MM, Mosely, RR. Length of stay in the VA: longterm care in short-term hospitals. Med Care 1987;25(3):250. 10. Fleming C, Fisher ES, Chang C, Bubolz TA, et al. Studying outcomes and hospital utilization in the elderly. Med Care 1992; 30(5):377. 11. Wolinsky FD, Coe RM, Mosely RR, et al. Veterans and nonveterans’ use of health services: a comparative analysis. Med Care 1985;23(12):1358. 12. Fleiss JL. Statistical methods for rates and proportions. 2nd ed. New York: Wiley, 1981. 13. SAS Institute. SAS/STAT user’s guide. Vol. 1 and 2. Version 6. Cary (NC): SAS Institute, 1990.