Preventive health performance rates in 2 clinic models—results of an intervention (before and after) in a resident practice (RP). NP indicates nurse practitioner practice; PSA, prostate-specific antigen determination.
Cardozo LJ, Steinberg J, Lepczyk MB, Binns-Emerick L, Cardozo Y, Aranha ANF. Improving Preventive Health Care in a Medical Resident Practice. Arch Intern Med. 1998;158(3):261-264. doi:10.1001/archinte.158.3.261
The ambulatory care resident practice is an opportunity to enhance the skills, knowledge, and attitudes for effective provision of preventive health services (PHS).
To determine whether a required intervention at a university medical resident practice would lead to improved performance of 6 secondary PHS.
A sequentially randomized chart analysis was performed at 2 clinics, a collaborative nurse practitioner practice (NP) and a resident practice (RP) to determine performance rate of secondary PHS (pelvic, prostate, and breast examinations, stool guaiac testing, mammography, and prostate-specific antigen determination). A significantly lower (36.9%) PHS performance rate was noted in the RP compared with 84.5% in NP for all 6 secondary PHS studied. An intervention was implemented in the RP: following every resident-patient clinic encounter a discussion and documentation of the patient's PHS status was required as part of the assessment and plans of management. At the end of 1 year the effect of this intervention on performance rates of the 6 PHS in the RP was analyzed.
There was a statistically significant difference (P<.001) between the PHS performance rates of NP and RP at the beginning. The intervention resulted in improved PHS performance rates in the RP; compared with the NP at the end of the study no statistical difference was noted between the groups.
Despite various task force recommendations, PHS performance rates are generally suboptimal in varied clinic settings, including those of resident practices in teaching hospitals. Physicians and residents believe in the importance of health maintenance but fall short of their ideal in practice. Interventions to improve performance rates have been described; we are detailing a simple, inexpensive, and practical method that achieved positive results.
IN 1990 the US Preventive Services Task Force1,2 recommended a list of primary and secondary preventive health services (PHS). While physicians support the concept of PHS, in practice they fall short of recommended national guidelines.3- 11 Existing outcome studies12,13 show compliance rates below expected levels in various practice models. Similar results have also been noted with internal medicine resident practices in teaching hospitals.14,15 There is further evidence that PHS outcomes are especially deficient in inner-city populations.16,17 This study compared PHS compliance rates in our resident practice (RP) with those of our collaborative nurse practitioner practice (NP), both serving inner-city patients. Results showed that the RP had significantly lower rates of provision of PHS. An intervention to improve provision of PHS in the RP was implemented, and at the end of 1 year, postintervention performances of 6 secondary PHS were analyzed.
This study was conducted at 2 sites in the Detroit Medical Center, Detroit, Mich, both of which provide care to a predominantly urban African American population. The NP is a geriatric clinic modeled after a collaborative practice style. It is staffed by 2 physicians and 3 nurse practitioners specializing in geriatric medicine. The clinic has approximately 900 active patients. Access to this clinic is by physician or self-referral and is restricted to patients older than 50 years. The NPs are available to their patients on a daily basis for any acute problem; however, routine care is rendered by appointment and provided primarily by the NP with physician supervision. In addition, every new patient seen by the NP has a follow-up visit with the physician. Patients are also seen by the physician if they are medically unstable, following an emergency department visit, after hospital discharge, and on a routine basis at least once a year.
The RP is located at the University Health Center, a part of the Detroit Medical Center, and serves as the general medicine RP where primary care is provided by medical residents who have their own group of patients. Residents typically spend half a day per week in the ambulatory care clinic and see their patients by appointment. However, patients with acute problems can access the clinic as necessary and are seen by a resident designated to take care of walk-in patients for that day. The RP is supervised by faculty physicians who discuss assessment and plans of treatment for every patient seen by the resident. There are a total of 800 patients who receive their primary care from the clinic.
The initial data collection involved a retrospective randomized chart analysis from the 2 ambulatory care clinics. A random sequential system to retrieve charts for study was used. Chart data abstraction was conducted only if the patient was older than 50 years and had received active ambulatory care during a 2-year period (1992-1994). Trained chart abstractors collected data on the sociodemographic variables, age, sex, marital status, health insurance, housing, education, and income levels of patients. Data were also collected on disease profile, number of drugs prescribed, and performance of 6 secondary PHS (pelvic examination and Papanicolaou smear, prostate and breast examinations, mammography, stool guaiac testing, and prostate-specific antigen determination). These were obtained from the problem list, medication list, preventive health maintenance list, and progress notes section of the clinic chart. A total of 132 patient charts in the NP group and 111 in the RP clinic were reviewed. No outcome hypothesis was made prior to the study, and the 2-year window (1992-1994) was considered adequate to allow for performance of the 6 secondary health maintenance activities studied. A random sample of 10 charts from the clinic were reabstracted to ensure interrater reliability.
Having documented a significantly lower (P<.001) PHS performance rate in the RP, we implemented an intervention to improve this process of care. The intervention required that following every resident-patient encounter, the resident also presented the current status of the patient's PHS to the attending physician. In addition, as part of every clinic visit progress note, documentation of the patient's PHS status was also required in the assessment and plans of treatment. Following the intervention period (1994-1995), a total of 100 charts from the RP and 116 charts from the NP were selected based on a random sequential system, and data on performance of the 6 secondary PHS were abstracted.
All data analysis and statistical procedures were conducted with the SPSS for Windows software package (release 6.0, SPSS Inc, Chicago, Ill). Covariates analyzed included age, sex, race, marital status, income, insurance coverage, and disease profile. Compliance was assessed by recording performance of preventive care services. The χ2 test was used to test and analyze categorical variables. Statistical significance on continuous variables was established using the Student t test. An α<.05 was considered significant.
Table 1 depicts patient characteristics in terms of age, sex, race, education, marital status, income, and health insurance coverage of the 2 clinic groups. Except for age (mean[±SD], 74.07±7.47 years in the RP compared with 77.39±7.73 years in the NP clinic), there were no statistical differences in the baseline characteristics of the 2 clinic populations.
The number of diseases per patient was statistically much higher in the RP clinic compared with the NP clinic (P<.001). This is a reflection of the pattern of referral of patients to the RP. Many patients are referred to the RP for outpatient follow-up, following hospitalization for acute medical problems. By contrast, patients in the NP clinic are either self-referred or physician-referred for longitudinal ambulatory care.
Figure 1 compares baseline compliance rates between the 2 sets of clinic patients in each of the 6 PHS areas studied. Overall, a significantly higher (χ2=29.68; P<.001) health maintenance performance rate was achieved in the NP clinic compared with those of the RP clinic. Following the intervention, there was a significant improvement in PHS in the RP clinic, and when the 2 practices were compared after intervention no statistical difference (χ2=3.27; P=.78) was noted in performance rates. Notably, pelvic examinations and Papanicolaou smears, and prostate and breast examinations showed the highest improvement rates in the RP. Results on prostate-specific antigen evaluation in both groups may reflect changes in recommendations and availability of the test during the study period.
Successful performance of preventive health care requires that physicians have appropriate knowledge, skills, and attitudes, that patients are well disposed to the concepts of preventive care, and that the practice structure facilitates this process of care. Among the many barriers to preventive health care, a low socioeconomic patient profile has been documented to contribute to suboptimal compliance with PHS.16- 19 While lack of insurance coverage may in some cases affect access to preventive care, studies in Canada (where universal coverage is available) still show a lower PHS compliance in patients from a lower socioeconomic status.20 It appears that having a regular source of primary care remains the single most important determinant of PHS performance,21 and this may account for why our clinics, both NP and RP (after intervention), achieved high secondary PHS performance despite serving patients with a low socioeconomic profile. Teaching hospitals are often the only primary care services able to provide PHS for the disadvantaged. However, some clinic locations within these large teaching hospitals have a complex operational structure, in addition the hospitals' focus on tertiary care, and their unfavorable patient demographics (elderly,22 minority, and inner-city populations16,17) may adversely affect PHS performance.21
Medical educators need, however, to recognize that a major determinant for achieving better PHS performance remains a physician's attitude and motivation toward the provision of preventive care.23 Traditional curricular emphasis at medical school on diagnosis of disease and treatment at the expense of disease prevention may negatively affect a physician's behavior toward prevention.24 Moreover, the lesser emphasis (at medical school) placed on epidemiology, biostatistics, behavioral, and social sciences is detrimental to highlighting the importance of preventive medicine. However, negative attitudes are further reinforced during residency when the acute care responsibilities of sicker patients leave residents with less time to focus on preventive medicine services in ambulatory care settings.24,25 A study by the American Cancer Society26 noted that physicians believed that their low PHS performance rates were attributable to lack of time or forgetfulness, patient refusal and discomfort with screening, or the logistics of patient access. In a survey on attitudes of house staff and attending physicians regarding health promotion and disease prevention in elderly individuals,22 lack of time was also perceived as the single most important factor. In addition, attending physicians identified suboptimal compensation and house staff lack of knowledge as secondary limiting factors in the provision of PHS. Dietrich and Goldberg27 documented that a physician's belief in the importance of PHS was a major determinant of performance; it was also noted that while house staff had more positive attitudes toward PHS, there appears, however, to be a lack of interest in preventive care by the clinical faculty. Some physicians expressed a lack of confidence in their ability to provide PHS whereas others viewed the role of preventive care medicine as inappropriate for physicians.
Headrick et al28 documented difficulties in securing medical resident compliance with therapeutic guidelines and national recommendations regarding cholesterol, despite receiving standard lectures. This problem is further compounded by the misperception resident physicians have that they provide much higher levels of PHS than they actually do.24,25 There have also been attempts to influence physician behavior and attitudes regarding PHS by using written materials, providing brief tutorials, having clinic chart "prompts" and "activated patients" willing to request PHS. Apart from the latter, the effects of these educational efforts in improving both provider and patient education on the importance of preventive care has yielded only limited improvement in performances.21,22
There have, however, been improved PHS performances following administrative and organizational changes regarding standing orders and provision for influenza vaccination,29 and the incorporation of clinical nurse practitioners into RPs.30,31 In addition computer-generated reminders, feedback from medical record data, and social influence–based methods that include academic detailing and expert reviews have improved PHS performances.32 We believe the ambulatory care experience during residency training is an ideal opportunity to develop positive attitudes toward preventive aspects of medical care. By requiring both discussion and documentation of the current status of PHS on every patient following every clinic visit, we noted a significant improvement in performance rates. Evaluation of a patient's preventive health care status at every clinic visit reinforces the importance of this aspect of health care to both the patient and resident physician. It also gives the attending physician a forum not only to emphasize the importance of preventive care but also to provide feedback relevant to the individual patient's needs.
The limitations of this study include the fact that it remains unclear if these positive PHS performance rates can be maintained on a long-term basis. In addition, our intervention was restricted to patients older than 50 years from a predominantly inner-city population of low socioeconomic means who received their care at a larger academic health center—its applicability to other communities and practice settings requires further study. Moreover, this intervention analysis targeted only the performance of 6 secondary PHS, while resident physician training must focus on all aspects of preventive care—primary, secondary, and tertiary. We also recognize that the 2 clinic settings are different. The NP is staffed by trained practitioners who focus on primary geriatric ambulatory care as opposed to residents in training who have other competing inpatient demands and a panel of sicker patients in their practice. However, despite these barriers the RP was able to achieve significant improvement in its preventive care performances after the intervention. Our results support previous findings33 that interventions designed to improve performances have a better chance of success if they are simple, practical, and inexpensive. From a teaching perspective, methods that use reinforcement techniques have been shown to be successful in enhancing knowledge, attitudes, and clinical practice skills.34,35 We recommend this intervention as a method to improve PHS performance in internal medicine resident clinic settings.
Accepted for publication June 27, 1997.
Presented in abstract form at the Eighth Biennial Teaching Internal Medicine Symposium, Seattle, Wash, September 1995.
We acknowledge Shlomo S. Sawilowsky, PhD, and Susan Eggly, MA, Wayne State University, for their review and helpful suggestions on the manuscript.
Corresponding author: Lavoisier J. Cardozo, MD, Division of General Internal Medicine, 5C, Wayne State University Health Center, 4201 St Antoine Dr, Detroit, MI 48201.