Journal of Community Health Vot. 15, No. 1, February 1990
PREVENTIVE CARE IN A VETERANS A D M I N I S T R A T I O N C O N T I N U I T Y CLINIC Alan Lefkowitz, MD, MS; Dorothy A. Snow, MD, MPH; Doris A. Cadigan, PhD
ABSTRACT: Low levels of compliance with established guidelines for preventive care have been documented in a variety of settings. AIthough the Veterans Administration (VA) is the largest health care provider in the United States, data concerning preventive care to veterans is lacking. A study was conducted to examine preventive care in a university affiliated VA continuity care clinic staffed by resident physicians. Recognized guidelines were used to generate criteria specifying appropriate periodic health examinations and preventive procedures. A r a n d o m sample of patients followed in the Baltimore VA Primary Care Clinic was reviewed. Compliance was highest for the traditional elements of the periodic health examination. Over 85% of patients had received histories and physical examinations, complete blood counts, electrolytes, chest radiographs, urinalyses, weight measurements, blood pressure determinations and electrocardiograms. Compliance was lower for cholesterol determination (51%) and for cancer screening procedures. O f appropriately selected patients, 19% had sigmoidoscopies, 44% rectal examinations, 50% PAP smears, and 17% mammograms. Lowest compliance was seen for influenza_immunization. Only 12.5% of high risk males and 0% of high risk females had received the vaccine. Compliance did not consistently vary by duration of clinic enrollment, age, race, or sex. Efforts must be made to improve compliance with established guidelines for preventive care.
The Veterans Administration (VA) is the largest health care system in the United States. Established in 1918 to provide inpatient care for service connected disabilities, the VA was authorized in 1973 to pro-
Alan Lefkowitz, MD, MS is Staff Physician Department of Internal Medicine, Johns Hopkins Health Plan Homewood Hospital Center, Baltimore, MD; Dorothy A. Snow, MD, MPH is Assistant Professor Department of Epidemiology and Preventive Medicine, Department of Internal Medicine University of Maryland School of Medicine and Associate Chief of Staff for Education, Baltimore Veterans Administration Medical Center; Doris A. Cadigan, PhD. at the time of this research was Research Associate Department of Epidemiology and Preventive Medicine University of Maryland School of Medicine, Baltimore, MD. Requests for reprints should be addressed to" Dorothy A. Snow, MD, MPH ACOS/Education (14A) VA Medical Center 3900 Loch Raven Blvd. Baltimore, MD 21218. Portions of this paper were presented at the Mid-Atlantic Regional Meeting of the Society of General Internal Medicine, Chevy Chase, Maryland, February 19, 1988 and also at "Prevention 88", Atlanta, Georgia, April 16, 1988. The authors wish to acknowledge Ms. Nancy Parks for her assistance in manuscript preparation and Ms. Wanda Shannon for her assistance in securing medical records. © 1990 Human Sciences Press
7
8
JOURNAL OF COMMUNITY HEALTH
vide "care on an ambulatory basis which would obviate the n e e d for hospitalization. ''~ Over 28 million U.S. veterans are now eligible for outpatient care, as well as hospitalization in VA facilities. Approximately 20 million ambulatory visits per year are provided in 172 medical centers and 229 outpatient facilities? In 1985, PL 98-160 authorized the VA to provide preventive health care to any veteran being treated by the VA? The VA Preventive Health Care Task Force identified a number of specific areas to be emphasized (see Table 1). Previous research suggests that patients receive fewer preventive services than is recommended by published guidelines. 47 Despite the size of the population served by the VA, little research has focused on this distinct patient pool. The purpose of this study was to investigate selected periodic health examinations and preventive procedures provided to patients followed in a university affiliated VA primary care setting. The project was conducted at the Primary Care Clinic (PCC) located at the Baltimore VA Medical Center. PCC provides continuity long term ambulatory care for veterans. The clinic is staffed by University of Maryland internal medicine resident physicians, nurse practitioners, and clinical pharmacists supervised by University of Maryland faculty. At the time of the research, 5 faculty and 31 resident physicians provided primary care to 1,017 patients. PCC patients have a mean age of 60 years and approximately 50% are white. Referrals to the PCC are from the acute outpatient area, subspecialty clinics, or after hospital discharge. Patients are assigned to a single provider, who acts as the coordinator of care. At the time of the study, care was provided at no cost to the patient. TABLE 1
Recommended Interventions Hypertension Screening/Treatment Alcohol Abuse Counseling Nutrition/Weight Control Counseling Smoking Cessation Counseling Physical Fitness/Exercise Programs Influenza Immunization Colorectal Cancer Screeing Diabetic Retinopathy Detection
Alan Lefkowitz et al.
9
METHODS
The study was a cross sectional survey of the records of patients enrolled in PCC. Study Population and Sample: The population included all veterans who were enrolled in and made one visit to the PCC during fiscal year 1987. The investigators drew a 14% random sample (one of every seventh chart from a randomly ordered clinic roster) of the male patients in the clinic. This yielded 144 patients. Exclusions were death (5), and non attendance (25) leaving 114 cases. Of these, 104 (91.2%) of charts were located for review. All 10 females in the clinic were also included, and will be analyzed separately. Data Collection: The entire inpatient and outpatient record for each patient was reviewed by an internist. Information collected included demographics, health care utilization, and the provision of periodic health examinations and preventive services. Criteria for Periodic Health Examinations and Preventive Services: Adaptations of the recommendations provided by Breslow and Somers, 8 the American Cancer Society9 (ACS), and the American College of Physicians ~°were used for the majority of the study criteria (Table 2). Criteria were considered to be met as follows: Medical History: documentation in the chart of a note with a chief complaint and 2 of the following 3 items: past history, social history, and, review of systems. Physical examination: documentation of a note including vital signs and 5 of the following examinations: fundoscopy; ear, nose, and throat; chest; cardiac; abdominal; genitourinary and neurologic. Laboratory Data, Other Procedures and Examinations: documented as completed, offered and refused, or done elsewhere. Smoking Cessation Counseling: documented in the progress notes. Criteria were considered met if delivered by the PCC or other health care provider. Statistical tests (student's t-test for differences between proportions) were performed to determine if the delivery of services varied by patient and provider characteristics.
RESULTS
Characteristics of Male Veterans: T h e male veterans h a d a m e a n age o f 62 years; 48% w e r e white. N e a r l y half o f the sample (47.5%) h a d b e e n hospitalized in the y e a r p r i o r to the study. T h e majority o f vete r a n s (68.3%) h a d b e e n a t t e n d i n g the clinic for b e t w e e n 1 a n d 5 years. O n average, veterans w e r e seen at the clinic every 115 days. T h e majority o f male v e t e r a n s (87%) h a d care p r o v i d e d by r e s i d e n t physicians. A
10
JOURNAL OF COMMUNITY H E A L T H
TABLE 2 Criteria for Screening Tests and Preventive Procedures a
Published Guidelines History & Physical Chest X-Ray
once every 5 years 8
EKG
once every 5 years 8
CBC
once every 5 years 8
Electrolytes
no indication
Cholesterol
once every 5 years 8
Urinalysis
no indication
Weight Check
every 2-3 years s
Blood Pressure
every 2-3 years (under 50) Every year (over 50) 8 annually (over 40) 9
Digital Rectal Examinition Stool Occult Blood Sigmoidoscopy
Ocular
Fundoscopic Examination Influenza Vaccination Smoking
no indication b
Study Criteria once every 5 years or since enrollment once every 5 years or since enrollment once every 5 years or since enrollment once every 5 years or since enrollment once every 5 years or since enrollment once every 5 years or since enrollment once every 5 years or since enrollment in last 2 years or since enrollment yearly for last 2 years (under 50) or once if enrolled for less t h a n 2 years (over 50) annually (over 40)
3 guaiac cards annually (over 50) 9 every 3-5 years after 2 annual negative tests (over 50) 9 annually 12 (for diabetics)
single guaiac card annually (over 50) once every 5 years or since enrollment (over 50)
annually for those over 65 and specified others ~° smoking history every five years 8
annually (over 65)
Annually (for diabetics)
counseling, or referral for counseling since enrollment
Alan Lefkowitz et al.
11
TABLE 2 (Continued) Mammography
Pap Smear
Every 1-2 years (under 49) annually (over 50) 9 every 3 years 9 after 2 annual negatives
Annually (over 50)
every 3 years
aUnless otherwise specified, guidelines are for adults aged 40-60. bAlthough not specifically recommended, data was collected because of the high prevalence of chest disease in the VA population 11
substantial proportion of the veterans (59.2%) were smokers, 32% of male veterans were diabetic. Delivery of Services to Male Veterans: Table 3 displays the percentages of male veterans receiving each service. A high percentage of men received services generally thought of as the traditional periodic examination. More than 85% of the male veterans received history and physical examinations, chest radiographs, electrocardiograms, complete blood counts, electrolytes, urinalyses, weight checks, and blood pressure monitoring within the specified time frame. Fifty one percent of the sample had cholesterol determined within the last five years. Screening examinations for colorectal cancer were performed less frequently. Of those veterans over 40, less than half (46.4%) received a digital rectal examination. Of those over 50, 44.2% had received an occult stool blood test, and 19% had received a sigmoidoscopy. Of the 33 diabetic veterans, 60% had received an ocular fundoscopy. With respect to influenza vaccinations, 12.5% of those over 65 received or were offered the service. Of smokers, 33.9% received smoking cessation counseling. The provision of services did not consistently vary by age, race, or length of clinic enrolhnent with two exceptions. Whites were more likely to have received influenza vaccine than non-whites (18.2 vs. 4.3% p < .05). Diabetic patients enrolled in clinic for more than two years were more likely to receive a retinopathy exam (77% vs. 31%; p < .01). Veterans cared for by resident physicians were more likely to receive rectal exams (50.6% vs. 16.9% p < .027) and stool occult blood determinations (48.8% vs. 15.4%; p < .024) than those having non-physician providers. Veterans not hospitalized in the past year were less likely than other veterans to have received a history and physical examination (77%
12
JOURNAL OF COMMUNITY HEALTH
TABLE 3 Delivery of Screening Procedures and Preventive Services to Male Veterans Enrolled in the Baltimore Veterans Administration Medical Center Primary Care Clinic (N = 104)
Procedure history and physical chest radiograph electrocardiogram complete blood count .electrolytes cholesterol determination urinalysis weight check blood pressure rectal exam (over age 40) stool occult blood (over age 50) sigmoidoscopy (over age 50) retinopathy exam (for diabetics) influenza vaccination (over age 65) smoking cessation
Time Interval for Screening
Percent Receiving Screening
5 years 5 years 5 years 5 years 5 years 5 years 5 years 2 years yearly × 2 yearly yearly
86.5 89.4 87.5 94.2 98.0 51.0 90.2 90.3 92.3 46.4 44.2
5 years
19.0
yearly
60.0
yearly
12.5
since enrollment
33.9
vs. 100%; p < .001), complete blood count (90.2% vs. 100%; p < .034), urinalysis (85% vs. 97.7%; p < .017), electrocardiogram (80.3% vs. 97.7%; p < .003) and rectal examinations (36.8% vs. 60%; p < .024). Although not statistically significant, veterans not hospitalized in the last year were also less likely to receive stool occult blood determinations (39% vs. 51%) and sigmoidoscopies (14% vs. 25%) than other veterans. Characteristics of Female Veterans: The ten women enrolled in the Primary Care Clinic had characteristics similar to the males. Their mean age was 52.7, almost 9 years younger than the men. Forty percent of the women were white. Ninety percent of the women had been hospitalized in the last five years. Like the men the majority of women (80%) had
Alan Lefkowitz et al.
13
been enrolled in the clinic for 1 to 5 years. The female veterans visited the clinic every 127 days, on average. Most (90%) had resident physicians as their providers. Like their male counterparts, 50% of the females were cigarette smokers. Three of the 10 women veterans were diabetic. Delivery of Services to Female Veterans: Table 4 displays the percent of female veterans receiving each service. At least 90% of the women received history and physical examinations, chest radiographs, complete blood counts, electrolytes, urinalyses, and blood pressure
TABLE 4
Delivery of Screening Procedures and Preventive Services to Female Veterans Enrolled in the Baltimore Veterans Administration Medical Center Primary Care Clinic (N = 10)
Procedure history and physical chest radiograph electrocardiogram complete blood count electrolytes cholesterol determination urinalysis weight check blood pressure rectal exam (over age 40) stool occult blood (over age 50) sigmoidoscopy (over age 50) retinopathy exam (for diabetics) influenza vaccination (over age 65) smoking cessation Pap smear Mammography
Time Interval for Screening 5 years 5 years 5 years 5 years 5 years 5 years 5 years 2 years yearly × 2 yearly yearly
Percent Receiving Screening 90.0 90.0 80.0 100.0 100:0 50.0 90.0 80.0 100.0 28.6 33.3
5 years
16.7
yearly
68.7
yearly
00.0
since enrollment every 3rd year annually
20.0 50.0 17.0
14
JOURNAL OF COMMUNITY HEALTH
checks. Half of the females had cholesterol d e t e r m i n e d within the last five years. Screening for colorectal cancer was as follows: of the 7 w o m e n who should have received a rectal exam, 2 (28.6%) did. Similarly, 2 (33.3%) of the 6 w o m e n over 50 had stool occult blood determinations, and 1 had a sigmoidoscopic examination. Neither of the 2 w o m e n over 65 had received or been offered an influenza vaccination. Twenty percent of the w o m e n smokers received some form of smoking cessation counseling. Of the 6 w o m e n over 50, one had received a m a m m o g r a m . Regarding cervical cancer screening, 4 of the veterans had hysterectomies. Of the remaining 6, 3 had received a pap smear within 3 years.
DISCUSSION This study examined the delivery of periodic health examinations and preventive ser.vices to veterans enrolled in the Baltimore VA Primary Care Clinic. Services that are often considered traditional components of the periodic health examination are likely to have been perf o r m e d on this g r o u p of patients. Procedures which are m o r e appropriately considered to be screening or preventive in nature such as cholesterol determination, influenza vaccination, smoking cessation counseling, and cancer screening are p e r f o r m e d at much less than optimal levels. While less than desirable these results approximate those f o u n d in other university settings and in the private sector (see Table 5). Although study methods differ, some comparisons may be made. Mandel et al. reviewed the medical records of a family medicine residents' practice. Over a five year period 53% of patients over 49 years of age received fecal occult blood testing, 19% of patients over 54 years of age received sigmoidoscopic examination, and 70% of w o m e n over 18 years of age received pap smears? 3 Reviewing an internal medicine faculty practice, McPhee et al. f o u n d that only 29% of patients had rectal examinations and stool occult blood tests in compliance with the ACS guidelines, and only 0.3% had a sigmoidoscopy. Compliance with other ACS recommendations was similarly low: 22% for pap smears and 13% for m a m m o g r a p h y . 7 Using ACS guidelines, a m o r e recent study in a family medicine residency p r o g r a m revealed the following rates of compliance: rectal examinations: 6% pap smears: 31.5%; m a m m o g r a p h y : 4%. Further only 6% had received cholesterol screening and 23% had received
Alan Lefkowitz et al.
15
TABLE 5
Comparison of Compliance with Preventive Screening Guidelines: Percent of Study Cases in Compliance with Study-Specific Guidelines
Procedure
Rectal Exam Stool Occult Blood Sigmoidoscopy Pap Smear Mammography Cholesterol Screening Influenza Vaccination Type of Provider
Setting
Baltimore VAMC
Mande113 et al, 1982
McPhee 7 et al, 1986
Morris & Morris/4 1988
46.4%
--
29.05%
6.0%
44.2% 19.0% 50.0% 17.0%
53.0% 19.0% 70.0% --
28.0% 0.3% 22.0% 13.0%
0.6% -31.5% 4.0%
51.0%
--
--
6.0%
12.5%
--
--
23.0%
Residents, Clinical Pharmacists, Nurse Practitioners Internal Medicine VAMC OutPatient Clinic
Residents
Residents, Residents Faculty, Nurse Practitioners
University Family Medicine Program Out-Patient Clinic
General Internal Medicine Out-Patient Practice
University Family Medicine Program Out-Patient Clinic
influenza vaccination?4 Lurie et al. studying care provided to enrollees of the Rand Health Insurance Experiment found rates of 57% for pap smears and 2% for mammograms. 6 Woo, et al. found that physicians performed cancer screenings such as rectal examinations, stool occult blood tests, sigmoidoscopies and mammograms less often than called for by physicians' own or published guidelines2 Across a variety of settings and types of providers, recognized guidelines for preventive health care fail to be met.
16
JOURNAL OF COMMUNITY HEALTH
It is likely that true preventive care in the current setting is provided at even lower rates than indicated. This study did not differentiate between procedures done for diagnosis and those done for screening purposes. Moreover, those patients who were not hospitalized within the previous year received fewer of the designated services, indicating that in many cases services may have been r e n d e r e d in pursuit of a diagnosis. Influenza vaccination represents a service which is always preventive in nature. In the current study, 12% of eligible male patients were offered or received influenza vaccination. Comparably, in a report of a faculty practice, Kosecoff et al. reported that 30% of eligible patients received an influenza vaccination in at least one of two seasons. 15 Fedson reported that without specific influenza programs, few patients receive vaccine. 16 T h e provision of services did not consistently vary by patient characteristics such as age, sex or race. Only for influenza vaccination were non-whites less likely to have received services. This is in contrast to data which imply that physicians providing care to minority patients are less likely to provide preventive servicesJ 7 Poor compliance with prevention guidelines may be due to a n u m b e r of factors including barriers involving patients, providers, and organizations. Despite advice from their personal physician, patients may be unwilling to undergo some preventive 'procedures. Cancer screening tests may heighten patient anxiety, making compliance less likely. Lay persons may be misinformed about the purpose of tests, women for example may not return for pelvic exams and pap smears once child bearing is complete. Patient discomfort or embarrassment may decrease cooperation with some procedures. It has been postulated that cost to the patient may contribute to the u n d e r performance of certain procedures. As care in the current study was provided at no cost to the patient, financial considerations probably do not explain the low rates of compliance seen. Physicians may lack the information, resources, or motivation to perform preventive services or they may disagree with the guidelines. In housestaff practice settings, it is unclear if lack of knowledge contributes to failure to provide r e c o m m e n d e d services. Woo, et al. found that residents overestimated the frequency with which screening should be performed. 4 T h e organization of medical care may bear a proportion of the blame for low rates of preventive care. Doctor-patient encounters may not afford time for preventive services. Some screening tests may re-
A l a n L e f k o w i t z et al,
17
quire facilities and personnel not easily accessible within the system. Even when tests appear accessible, there may be significant barriers created by lack of sufficient ancillary workers to streamline the process of performing or ordering a test. In summary, the study results indicate that in this setting compliance with traditional components of the periodic health examination is relatively high, but that provision of cancer screening and other preventive services is quite low. This report examines only one clinic, and there may be considerable variability among VA settings. However, it is very likely that this pattern is seen throughout the VA system as it is in other health care organizations. The VA has taken recent steps to emphasize and improve preventive care. Since these efforts began in earnest only recently, any resulting improvements are not reflected in the current study. Programs and efforts to achieve improved preventive care can be of value. Prior to such interventions it is important to establish baseline data to allow the evaluation of these efforts. Given the large number of patients potentially effected by even minor improvements in prevention, it is imperative that efforts continue to improve the quality of health care delivery to the veteran population. REFERENCES 1. Public Law 93-82 Veterans Health Care Expansion Act of 1973, August 2, 1973. 2. United States Veterans Administration, Administrator of Veterans Affairs: Annual Report 1985. Washington, DC Veterans Administration. 3. Public Law 98-160 Veterans Administration Health-Care Programs, November 21, 1983. 4. Woo B, Woo B, Cook EF, et al: Screening procedures in the asymptomatic adult: comparison of physicians' recommendations, patient's decisions, published guidelines, and actual practice. JAMA 254:1480-1484, 1985. 5. Cohen DI, Littenberg B, Wetzel C, et al: Improving physician compliance with preventive guidelines. Med Care 20:1040-1045, 1982. 6. Lurie N, Manning WG, PetersonC, et al: Preventive care: do we practice what we preach? Am J Public Health 77:801-804, 1987. 7. McPhee SJ, Richard RJ, Solkowitz SN: Performance of cancer screening in a university general internal medicine practice: comparison with the 1980 American Cancer Society guidelines. J Gen Int Med 1:275-281, 1986. 8. Breslow L, Somers AR: The lifetime health monitoring program: A practical approach to preventive medicine. N EngJ Med 296:601-608, 1977. 9. American Cancer Society: Guidelines for the cancer related checkup. CA 30:194-240, 1980. 10. American College of Physicians Committee on Immunization: Guidefor Adult Immunizations. Philadelphia: American College of Physicians p. 58, 1985. 11. Fink DJ, Fang M, Wyle FA: Routine chest x-ray films in a veterans hospital. JAMA 245:1056-1057, 1981. 12. Marble A (ed): Joslin's Diabetes Mellitus. Philadelphia: Lea & Feibiger p. 618, 1985. 13. Mandel IG, Franks P, Dickinson JC: Screening guidelines in a family medicine program: A five year experience. J Fam Practice 14:901-907, 1982. 14. Morris PD, Morris ER: Family practice residents' compliance with preventive medicine recommendations. Am J Prev Med 4:161-165, 1988.
18
JOURNAL OF COMMUNITY HEALTH
15. KosecoffJ, et al: General medicine care and the education of internists in university hospitals. Ann Intern Med 102:250-257, 1985. 16. Fedson DS: Influenza and pneumococcal immunization in medical clinics, 1971-1983.Jlnfect Dis 149:817-818, 1984. 17. Gemson DH, EUison J, Messeri P: Differences in physician prevention practice patterns for white and minority patients. J Comm Health 13:53-64, 1988.
PREVENTIVE CARE IN A VETERANS A D M I N I S T R A T I O N C O N T I N U I T Y CLINIC Alan Lefkowitz, MD, MS; Dorothy A. Snow, MD, MPH; Doris A. Cadigan, PhD
ABSTRACT: Low levels of compliance with established guidelines for preventive care have been documented in a variety of settings. AIthough the Veterans Administration (VA) is the largest health care provider in the United States, data concerning preventive care to veterans is lacking. A study was conducted to examine preventive care in a university affiliated VA continuity care clinic staffed by resident physicians. Recognized guidelines were used to generate criteria specifying appropriate periodic health examinations and preventive procedures. A r a n d o m sample of patients followed in the Baltimore VA Primary Care Clinic was reviewed. Compliance was highest for the traditional elements of the periodic health examination. Over 85% of patients had received histories and physical examinations, complete blood counts, electrolytes, chest radiographs, urinalyses, weight measurements, blood pressure determinations and electrocardiograms. Compliance was lower for cholesterol determination (51%) and for cancer screening procedures. O f appropriately selected patients, 19% had sigmoidoscopies, 44% rectal examinations, 50% PAP smears, and 17% mammograms. Lowest compliance was seen for influenza_immunization. Only 12.5% of high risk males and 0% of high risk females had received the vaccine. Compliance did not consistently vary by duration of clinic enrollment, age, race, or sex. Efforts must be made to improve compliance with established guidelines for preventive care.
The Veterans Administration (VA) is the largest health care system in the United States. Established in 1918 to provide inpatient care for service connected disabilities, the VA was authorized in 1973 to pro-
Alan Lefkowitz, MD, MS is Staff Physician Department of Internal Medicine, Johns Hopkins Health Plan Homewood Hospital Center, Baltimore, MD; Dorothy A. Snow, MD, MPH is Assistant Professor Department of Epidemiology and Preventive Medicine, Department of Internal Medicine University of Maryland School of Medicine and Associate Chief of Staff for Education, Baltimore Veterans Administration Medical Center; Doris A. Cadigan, PhD. at the time of this research was Research Associate Department of Epidemiology and Preventive Medicine University of Maryland School of Medicine, Baltimore, MD. Requests for reprints should be addressed to" Dorothy A. Snow, MD, MPH ACOS/Education (14A) VA Medical Center 3900 Loch Raven Blvd. Baltimore, MD 21218. Portions of this paper were presented at the Mid-Atlantic Regional Meeting of the Society of General Internal Medicine, Chevy Chase, Maryland, February 19, 1988 and also at "Prevention 88", Atlanta, Georgia, April 16, 1988. The authors wish to acknowledge Ms. Nancy Parks for her assistance in manuscript preparation and Ms. Wanda Shannon for her assistance in securing medical records. © 1990 Human Sciences Press
7
8
JOURNAL OF COMMUNITY HEALTH
vide "care on an ambulatory basis which would obviate the n e e d for hospitalization. ''~ Over 28 million U.S. veterans are now eligible for outpatient care, as well as hospitalization in VA facilities. Approximately 20 million ambulatory visits per year are provided in 172 medical centers and 229 outpatient facilities? In 1985, PL 98-160 authorized the VA to provide preventive health care to any veteran being treated by the VA? The VA Preventive Health Care Task Force identified a number of specific areas to be emphasized (see Table 1). Previous research suggests that patients receive fewer preventive services than is recommended by published guidelines. 47 Despite the size of the population served by the VA, little research has focused on this distinct patient pool. The purpose of this study was to investigate selected periodic health examinations and preventive procedures provided to patients followed in a university affiliated VA primary care setting. The project was conducted at the Primary Care Clinic (PCC) located at the Baltimore VA Medical Center. PCC provides continuity long term ambulatory care for veterans. The clinic is staffed by University of Maryland internal medicine resident physicians, nurse practitioners, and clinical pharmacists supervised by University of Maryland faculty. At the time of the research, 5 faculty and 31 resident physicians provided primary care to 1,017 patients. PCC patients have a mean age of 60 years and approximately 50% are white. Referrals to the PCC are from the acute outpatient area, subspecialty clinics, or after hospital discharge. Patients are assigned to a single provider, who acts as the coordinator of care. At the time of the study, care was provided at no cost to the patient. TABLE 1
Recommended Interventions Hypertension Screening/Treatment Alcohol Abuse Counseling Nutrition/Weight Control Counseling Smoking Cessation Counseling Physical Fitness/Exercise Programs Influenza Immunization Colorectal Cancer Screeing Diabetic Retinopathy Detection
Alan Lefkowitz et al.
9
METHODS
The study was a cross sectional survey of the records of patients enrolled in PCC. Study Population and Sample: The population included all veterans who were enrolled in and made one visit to the PCC during fiscal year 1987. The investigators drew a 14% random sample (one of every seventh chart from a randomly ordered clinic roster) of the male patients in the clinic. This yielded 144 patients. Exclusions were death (5), and non attendance (25) leaving 114 cases. Of these, 104 (91.2%) of charts were located for review. All 10 females in the clinic were also included, and will be analyzed separately. Data Collection: The entire inpatient and outpatient record for each patient was reviewed by an internist. Information collected included demographics, health care utilization, and the provision of periodic health examinations and preventive services. Criteria for Periodic Health Examinations and Preventive Services: Adaptations of the recommendations provided by Breslow and Somers, 8 the American Cancer Society9 (ACS), and the American College of Physicians ~°were used for the majority of the study criteria (Table 2). Criteria were considered to be met as follows: Medical History: documentation in the chart of a note with a chief complaint and 2 of the following 3 items: past history, social history, and, review of systems. Physical examination: documentation of a note including vital signs and 5 of the following examinations: fundoscopy; ear, nose, and throat; chest; cardiac; abdominal; genitourinary and neurologic. Laboratory Data, Other Procedures and Examinations: documented as completed, offered and refused, or done elsewhere. Smoking Cessation Counseling: documented in the progress notes. Criteria were considered met if delivered by the PCC or other health care provider. Statistical tests (student's t-test for differences between proportions) were performed to determine if the delivery of services varied by patient and provider characteristics.
RESULTS
Characteristics of Male Veterans: T h e male veterans h a d a m e a n age o f 62 years; 48% w e r e white. N e a r l y half o f the sample (47.5%) h a d b e e n hospitalized in the y e a r p r i o r to the study. T h e majority o f vete r a n s (68.3%) h a d b e e n a t t e n d i n g the clinic for b e t w e e n 1 a n d 5 years. O n average, veterans w e r e seen at the clinic every 115 days. T h e majority o f male v e t e r a n s (87%) h a d care p r o v i d e d by r e s i d e n t physicians. A
10
JOURNAL OF COMMUNITY H E A L T H
TABLE 2 Criteria for Screening Tests and Preventive Procedures a
Published Guidelines History & Physical Chest X-Ray
once every 5 years 8
EKG
once every 5 years 8
CBC
once every 5 years 8
Electrolytes
no indication
Cholesterol
once every 5 years 8
Urinalysis
no indication
Weight Check
every 2-3 years s
Blood Pressure
every 2-3 years (under 50) Every year (over 50) 8 annually (over 40) 9
Digital Rectal Examinition Stool Occult Blood Sigmoidoscopy
Ocular
Fundoscopic Examination Influenza Vaccination Smoking
no indication b
Study Criteria once every 5 years or since enrollment once every 5 years or since enrollment once every 5 years or since enrollment once every 5 years or since enrollment once every 5 years or since enrollment once every 5 years or since enrollment once every 5 years or since enrollment in last 2 years or since enrollment yearly for last 2 years (under 50) or once if enrolled for less t h a n 2 years (over 50) annually (over 40)
3 guaiac cards annually (over 50) 9 every 3-5 years after 2 annual negative tests (over 50) 9 annually 12 (for diabetics)
single guaiac card annually (over 50) once every 5 years or since enrollment (over 50)
annually for those over 65 and specified others ~° smoking history every five years 8
annually (over 65)
Annually (for diabetics)
counseling, or referral for counseling since enrollment
Alan Lefkowitz et al.
11
TABLE 2 (Continued) Mammography
Pap Smear
Every 1-2 years (under 49) annually (over 50) 9 every 3 years 9 after 2 annual negatives
Annually (over 50)
every 3 years
aUnless otherwise specified, guidelines are for adults aged 40-60. bAlthough not specifically recommended, data was collected because of the high prevalence of chest disease in the VA population 11
substantial proportion of the veterans (59.2%) were smokers, 32% of male veterans were diabetic. Delivery of Services to Male Veterans: Table 3 displays the percentages of male veterans receiving each service. A high percentage of men received services generally thought of as the traditional periodic examination. More than 85% of the male veterans received history and physical examinations, chest radiographs, electrocardiograms, complete blood counts, electrolytes, urinalyses, weight checks, and blood pressure monitoring within the specified time frame. Fifty one percent of the sample had cholesterol determined within the last five years. Screening examinations for colorectal cancer were performed less frequently. Of those veterans over 40, less than half (46.4%) received a digital rectal examination. Of those over 50, 44.2% had received an occult stool blood test, and 19% had received a sigmoidoscopy. Of the 33 diabetic veterans, 60% had received an ocular fundoscopy. With respect to influenza vaccinations, 12.5% of those over 65 received or were offered the service. Of smokers, 33.9% received smoking cessation counseling. The provision of services did not consistently vary by age, race, or length of clinic enrolhnent with two exceptions. Whites were more likely to have received influenza vaccine than non-whites (18.2 vs. 4.3% p < .05). Diabetic patients enrolled in clinic for more than two years were more likely to receive a retinopathy exam (77% vs. 31%; p < .01). Veterans cared for by resident physicians were more likely to receive rectal exams (50.6% vs. 16.9% p < .027) and stool occult blood determinations (48.8% vs. 15.4%; p < .024) than those having non-physician providers. Veterans not hospitalized in the past year were less likely than other veterans to have received a history and physical examination (77%
12
JOURNAL OF COMMUNITY HEALTH
TABLE 3 Delivery of Screening Procedures and Preventive Services to Male Veterans Enrolled in the Baltimore Veterans Administration Medical Center Primary Care Clinic (N = 104)
Procedure history and physical chest radiograph electrocardiogram complete blood count .electrolytes cholesterol determination urinalysis weight check blood pressure rectal exam (over age 40) stool occult blood (over age 50) sigmoidoscopy (over age 50) retinopathy exam (for diabetics) influenza vaccination (over age 65) smoking cessation
Time Interval for Screening
Percent Receiving Screening
5 years 5 years 5 years 5 years 5 years 5 years 5 years 2 years yearly × 2 yearly yearly
86.5 89.4 87.5 94.2 98.0 51.0 90.2 90.3 92.3 46.4 44.2
5 years
19.0
yearly
60.0
yearly
12.5
since enrollment
33.9
vs. 100%; p < .001), complete blood count (90.2% vs. 100%; p < .034), urinalysis (85% vs. 97.7%; p < .017), electrocardiogram (80.3% vs. 97.7%; p < .003) and rectal examinations (36.8% vs. 60%; p < .024). Although not statistically significant, veterans not hospitalized in the last year were also less likely to receive stool occult blood determinations (39% vs. 51%) and sigmoidoscopies (14% vs. 25%) than other veterans. Characteristics of Female Veterans: The ten women enrolled in the Primary Care Clinic had characteristics similar to the males. Their mean age was 52.7, almost 9 years younger than the men. Forty percent of the women were white. Ninety percent of the women had been hospitalized in the last five years. Like the men the majority of women (80%) had
Alan Lefkowitz et al.
13
been enrolled in the clinic for 1 to 5 years. The female veterans visited the clinic every 127 days, on average. Most (90%) had resident physicians as their providers. Like their male counterparts, 50% of the females were cigarette smokers. Three of the 10 women veterans were diabetic. Delivery of Services to Female Veterans: Table 4 displays the percent of female veterans receiving each service. At least 90% of the women received history and physical examinations, chest radiographs, complete blood counts, electrolytes, urinalyses, and blood pressure
TABLE 4
Delivery of Screening Procedures and Preventive Services to Female Veterans Enrolled in the Baltimore Veterans Administration Medical Center Primary Care Clinic (N = 10)
Procedure history and physical chest radiograph electrocardiogram complete blood count electrolytes cholesterol determination urinalysis weight check blood pressure rectal exam (over age 40) stool occult blood (over age 50) sigmoidoscopy (over age 50) retinopathy exam (for diabetics) influenza vaccination (over age 65) smoking cessation Pap smear Mammography
Time Interval for Screening 5 years 5 years 5 years 5 years 5 years 5 years 5 years 2 years yearly × 2 yearly yearly
Percent Receiving Screening 90.0 90.0 80.0 100.0 100:0 50.0 90.0 80.0 100.0 28.6 33.3
5 years
16.7
yearly
68.7
yearly
00.0
since enrollment every 3rd year annually
20.0 50.0 17.0
14
JOURNAL OF COMMUNITY HEALTH
checks. Half of the females had cholesterol d e t e r m i n e d within the last five years. Screening for colorectal cancer was as follows: of the 7 w o m e n who should have received a rectal exam, 2 (28.6%) did. Similarly, 2 (33.3%) of the 6 w o m e n over 50 had stool occult blood determinations, and 1 had a sigmoidoscopic examination. Neither of the 2 w o m e n over 65 had received or been offered an influenza vaccination. Twenty percent of the w o m e n smokers received some form of smoking cessation counseling. Of the 6 w o m e n over 50, one had received a m a m m o g r a m . Regarding cervical cancer screening, 4 of the veterans had hysterectomies. Of the remaining 6, 3 had received a pap smear within 3 years.
DISCUSSION This study examined the delivery of periodic health examinations and preventive ser.vices to veterans enrolled in the Baltimore VA Primary Care Clinic. Services that are often considered traditional components of the periodic health examination are likely to have been perf o r m e d on this g r o u p of patients. Procedures which are m o r e appropriately considered to be screening or preventive in nature such as cholesterol determination, influenza vaccination, smoking cessation counseling, and cancer screening are p e r f o r m e d at much less than optimal levels. While less than desirable these results approximate those f o u n d in other university settings and in the private sector (see Table 5). Although study methods differ, some comparisons may be made. Mandel et al. reviewed the medical records of a family medicine residents' practice. Over a five year period 53% of patients over 49 years of age received fecal occult blood testing, 19% of patients over 54 years of age received sigmoidoscopic examination, and 70% of w o m e n over 18 years of age received pap smears? 3 Reviewing an internal medicine faculty practice, McPhee et al. f o u n d that only 29% of patients had rectal examinations and stool occult blood tests in compliance with the ACS guidelines, and only 0.3% had a sigmoidoscopy. Compliance with other ACS recommendations was similarly low: 22% for pap smears and 13% for m a m m o g r a p h y . 7 Using ACS guidelines, a m o r e recent study in a family medicine residency p r o g r a m revealed the following rates of compliance: rectal examinations: 6% pap smears: 31.5%; m a m m o g r a p h y : 4%. Further only 6% had received cholesterol screening and 23% had received
Alan Lefkowitz et al.
15
TABLE 5
Comparison of Compliance with Preventive Screening Guidelines: Percent of Study Cases in Compliance with Study-Specific Guidelines
Procedure
Rectal Exam Stool Occult Blood Sigmoidoscopy Pap Smear Mammography Cholesterol Screening Influenza Vaccination Type of Provider
Setting
Baltimore VAMC
Mande113 et al, 1982
McPhee 7 et al, 1986
Morris & Morris/4 1988
46.4%
--
29.05%
6.0%
44.2% 19.0% 50.0% 17.0%
53.0% 19.0% 70.0% --
28.0% 0.3% 22.0% 13.0%
0.6% -31.5% 4.0%
51.0%
--
--
6.0%
12.5%
--
--
23.0%
Residents, Clinical Pharmacists, Nurse Practitioners Internal Medicine VAMC OutPatient Clinic
Residents
Residents, Residents Faculty, Nurse Practitioners
University Family Medicine Program Out-Patient Clinic
General Internal Medicine Out-Patient Practice
University Family Medicine Program Out-Patient Clinic
influenza vaccination?4 Lurie et al. studying care provided to enrollees of the Rand Health Insurance Experiment found rates of 57% for pap smears and 2% for mammograms. 6 Woo, et al. found that physicians performed cancer screenings such as rectal examinations, stool occult blood tests, sigmoidoscopies and mammograms less often than called for by physicians' own or published guidelines2 Across a variety of settings and types of providers, recognized guidelines for preventive health care fail to be met.
16
JOURNAL OF COMMUNITY HEALTH
It is likely that true preventive care in the current setting is provided at even lower rates than indicated. This study did not differentiate between procedures done for diagnosis and those done for screening purposes. Moreover, those patients who were not hospitalized within the previous year received fewer of the designated services, indicating that in many cases services may have been r e n d e r e d in pursuit of a diagnosis. Influenza vaccination represents a service which is always preventive in nature. In the current study, 12% of eligible male patients were offered or received influenza vaccination. Comparably, in a report of a faculty practice, Kosecoff et al. reported that 30% of eligible patients received an influenza vaccination in at least one of two seasons. 15 Fedson reported that without specific influenza programs, few patients receive vaccine. 16 T h e provision of services did not consistently vary by patient characteristics such as age, sex or race. Only for influenza vaccination were non-whites less likely to have received services. This is in contrast to data which imply that physicians providing care to minority patients are less likely to provide preventive servicesJ 7 Poor compliance with prevention guidelines may be due to a n u m b e r of factors including barriers involving patients, providers, and organizations. Despite advice from their personal physician, patients may be unwilling to undergo some preventive 'procedures. Cancer screening tests may heighten patient anxiety, making compliance less likely. Lay persons may be misinformed about the purpose of tests, women for example may not return for pelvic exams and pap smears once child bearing is complete. Patient discomfort or embarrassment may decrease cooperation with some procedures. It has been postulated that cost to the patient may contribute to the u n d e r performance of certain procedures. As care in the current study was provided at no cost to the patient, financial considerations probably do not explain the low rates of compliance seen. Physicians may lack the information, resources, or motivation to perform preventive services or they may disagree with the guidelines. In housestaff practice settings, it is unclear if lack of knowledge contributes to failure to provide r e c o m m e n d e d services. Woo, et al. found that residents overestimated the frequency with which screening should be performed. 4 T h e organization of medical care may bear a proportion of the blame for low rates of preventive care. Doctor-patient encounters may not afford time for preventive services. Some screening tests may re-
A l a n L e f k o w i t z et al,
17
quire facilities and personnel not easily accessible within the system. Even when tests appear accessible, there may be significant barriers created by lack of sufficient ancillary workers to streamline the process of performing or ordering a test. In summary, the study results indicate that in this setting compliance with traditional components of the periodic health examination is relatively high, but that provision of cancer screening and other preventive services is quite low. This report examines only one clinic, and there may be considerable variability among VA settings. However, it is very likely that this pattern is seen throughout the VA system as it is in other health care organizations. The VA has taken recent steps to emphasize and improve preventive care. Since these efforts began in earnest only recently, any resulting improvements are not reflected in the current study. Programs and efforts to achieve improved preventive care can be of value. Prior to such interventions it is important to establish baseline data to allow the evaluation of these efforts. Given the large number of patients potentially effected by even minor improvements in prevention, it is imperative that efforts continue to improve the quality of health care delivery to the veteran population. REFERENCES 1. Public Law 93-82 Veterans Health Care Expansion Act of 1973, August 2, 1973. 2. United States Veterans Administration, Administrator of Veterans Affairs: Annual Report 1985. Washington, DC Veterans Administration. 3. Public Law 98-160 Veterans Administration Health-Care Programs, November 21, 1983. 4. Woo B, Woo B, Cook EF, et al: Screening procedures in the asymptomatic adult: comparison of physicians' recommendations, patient's decisions, published guidelines, and actual practice. JAMA 254:1480-1484, 1985. 5. Cohen DI, Littenberg B, Wetzel C, et al: Improving physician compliance with preventive guidelines. Med Care 20:1040-1045, 1982. 6. Lurie N, Manning WG, PetersonC, et al: Preventive care: do we practice what we preach? Am J Public Health 77:801-804, 1987. 7. McPhee SJ, Richard RJ, Solkowitz SN: Performance of cancer screening in a university general internal medicine practice: comparison with the 1980 American Cancer Society guidelines. J Gen Int Med 1:275-281, 1986. 8. Breslow L, Somers AR: The lifetime health monitoring program: A practical approach to preventive medicine. N EngJ Med 296:601-608, 1977. 9. American Cancer Society: Guidelines for the cancer related checkup. CA 30:194-240, 1980. 10. American College of Physicians Committee on Immunization: Guidefor Adult Immunizations. Philadelphia: American College of Physicians p. 58, 1985. 11. Fink DJ, Fang M, Wyle FA: Routine chest x-ray films in a veterans hospital. JAMA 245:1056-1057, 1981. 12. Marble A (ed): Joslin's Diabetes Mellitus. Philadelphia: Lea & Feibiger p. 618, 1985. 13. Mandel IG, Franks P, Dickinson JC: Screening guidelines in a family medicine program: A five year experience. J Fam Practice 14:901-907, 1982. 14. Morris PD, Morris ER: Family practice residents' compliance with preventive medicine recommendations. Am J Prev Med 4:161-165, 1988.
18
JOURNAL OF COMMUNITY HEALTH
15. KosecoffJ, et al: General medicine care and the education of internists in university hospitals. Ann Intern Med 102:250-257, 1985. 16. Fedson DS: Influenza and pneumococcal immunization in medical clinics, 1971-1983.Jlnfect Dis 149:817-818, 1984. 17. Gemson DH, EUison J, Messeri P: Differences in physician prevention practice patterns for white and minority patients. J Comm Health 13:53-64, 1988.
