Pourat, Nadereh PhD*†; Kominski, Gerald F. PhD*†; Nihalani, Jas MPH‡; Neiman, Romni BS§; Bolan, Gail MD§
An estimated 2.8 million Americans are infected with Chlamydia trachomatis (CT) each year, which is identified as a “silent epidemic.”1 A readily treated bacterial infection with potentially serious complications for untreated disease, CT control efforts focus on mass screening of young females who have the highest incidence of disease. The Centers for Disease Control and Prevention (CDC) has recommended CT screening of sexually active young women since 1993.2–4 In 2000, CT screening was also added as a Healthcare Effectiveness Data and Information Set (HEDIS) performance measure, providing strong incentives for health maintenance organizations (HMOs) to increase CT screening rates among their enrollees.
Medicaid HMOs may have stronger incentives to improve CT screening rates given the high proportion of young females, and therefore high risk, among Medicaid beneficiaries.5–8 Between 2000 and 2008, the proportion of Medicaid enrollees screened for CT increased (from 37% to 53% for ages 16–20 and 21% to 59% for ages 21–24) more than commercial HMOs (30% to 40% for ages 16–20 and 29% to 44% for ages 21–24).9,10 Despite stronger incentives and better performance than commercial HMOs, data from a 1999 study showed that some Medicaid HMOs did not prioritize sexually transmitted disease (STD) promotion and control and did not consistently recommend compliance with CT and other STD guidelines.7
HMO efforts to influence physician practice may be achieved by recommending guidelines and other interventions, including physician education, review of physician practices, and providing feedback.11,12 More recently, financial incentives have been used to influence physician practice, for example, by including such incentives in contractual agreements or providing pay for performance.13–15 These interventions are perceived to improve provider awareness of and compliance with best clinical practices under the general umbrella of quality control efforts in managed care.15–17
Influencing physician behavior is challenging, for Medicaid HMOs and other organized delivery systems including medical groups and state and local agencies. Barriers to physician adherence with guidelines include lack of awareness and familiarity, lack of agreement, lack of outcome expectancy, lack of time, difficulty of use, difficulty in achieving patient compliance, and numerous organizational constraints.18,19 Physician adherence may be negatively affected if guidelines are perceived to be motivated primarily to reduce costs or to punish poor performance.17,20 In most instances, physicians must view guidelines as useful to comply with them.21 Compliance is particularly complicated when physicians have multiple HMO or medical group contracts and have to process the information provided from multiple sources and face potentially conflicting financial or nonfinancial incentives. Multiple mandates from HMOs have been a barrier to physician delivery of STD care.22 Compliance with guidelines also varies by physician gender, specialty, and years of practice.23,24
Few studies have examined physician adherence with CT screening and management guidelines in managed care organizations. Although adherence with CDC guidelines for the treatment of CT in 2 HMOs of California was high in 1 study, physician adherence with screening guidelines in selected HMOs nationwide was low in another.7,25 In 1998, a study of 3 HMOs in Washington found CT testing for less than one-third of sexually active young females.26 No studies to date have examined the independent relationship of HMO interventions with physician CT screening practices, which is the focus of this study.
We examine the relationship of HMO interventions such as recommendations for physicians to annually screen young sexually active females, dissemination of CT screening guidelines to physicians by mail, review of CT screening practices and/or feedback to physicians, and inclusion of quality of care and profit sharing in payment criteria. In this study, we hypothesized that physicians would be more likely to consistently adhere to CT screening guidelines if a higher percent of their contracted Medicaid HMOs recommended these guidelines, disseminated it, trained physicians, reviewed physician screening practices and provided feedback, or included quality of care as a payment criterion. Alternatively, inclusion of profit sharing as a payment criterion was expected to reduce the likelihood of adherence, since annual screening would increase utilization of services, and thus increase costs without increasing payment.
Data and Sample
Data were obtained from a 2002 survey of 17 Medicaid managed care health plans and 941 of their contracted physicians in 8 California counties, with the highest rates of both CT infections and the highest number of Medicaid HMO enrollees in 2002. These counties accounted for 71% of CT cases, 63% of Medicaid HMO enrollees, and 65% of California's population. The medical directors of Medicaid HMOs operating in these counties completed a 10-minute, self-administered survey, whereas 2 of them with less than 10,000 enrollees each refused to participate in the study. The respondents answered questions about HMO interventions directed at physician adherence to CT screening guidelines. A database of all participating physicians was constructed from an electronic version of the participating plans' PCP directories and included physicians in general practice, family practice, internal medicine, pediatrics, and obstetrics/gynecology, who acted as primary care physicians. A random sample of physicians was contacted up to 12 times for 15-minute phone interviews from January through May 2002 with a $75 financial incentive. Physicians were surveyed on their CT and other STD screening and treatment practices as well as practice and business characteristics. The adjusted response rate was 41%. Of the list of physicians provided by participating HMOs, 52% (4932) were found to be ineligible because of outdated contact information including contractual changes in the time between the collection of the primary care physicians list from HMOs and the fielding of the survey and also due to being specialists outside the scope of the study. Another 18% (the population not contacted—called multiple times according to survey protocol but never refused or provided any information on eligibility) were estimated to be ineligible following the methodology used in another national survey of physicians and approved by the American Association for Public Opinion Research.27,28 Additional analysis of respondents and nonrespondents on the basis of available characteristics, including county and specialty, did not identify any nonresponse bias.13,29 This study was approved by the appropriate institutional review board, and all subjects of the study consented to participate in the survey verbally at the time of interview.
Annual CT screening of sexually active females between ages 15 and 25 years is recommended by the CDC and US Preventive Services Task Force (USPSTF).4,30 Annual CT screening of women aged 15 to 25 years was divided into 2 variables, because pediatricians were included in the survey and would infrequently see patients older than 19 years of age.
Physician adherence with this guideline was captured in the following 2 questions: “Do you test sexually active adolescent females (ages, 15–19) for Chlamydia annually?” and “Do you test sexually active young women (ages, 20–25) for Chlamydia annually?” The response categories ranged from 1 to 5 on a 5-point Likert scale, with “always” (1), “usually” (2), “sometimes” (3), “rarely” (4), and “never” (5). Consistent adherence was defined as “always” and “usually,” following best practices because adherence to any practice guidelines often depends on the physician judgment of the appropriateness of the treatment or procedure given the presentation of illness, the patient's characteristics, and other circumstances.
HMOs reported on whether they explicitly recommended annual CT screening of young females, disseminated STD clinical management and prevention guidelines by mail, provided STD guideline training to contracted physicians, reviewed physicians' STD screening practices and provided feedback on those practices, and included quality of care (e.g., patient satisfaction or peer review) and profit sharing as criteria for payment in physician contracts. Each physician named up to 7 contracted Medicaid HMOs, resulting in up to 7 recommendations/other interventions from different Medicaid HMOs for each physician. The physicians were not asked to report on more than 7 contracted HMOs because of space limitations. For each care management practice, we calculated the percentage of Medicaid HMOs per each physician who reported such practices.
Review and feedback by HMOs were highly correlated because many HMOs conducted either both or neither activity. Thus, a 3-category variable was created for this concept, identifying HMOs as follows: neither reviewed nor provided feedback, reviewed but did not provide feedback, and both reviewed and provided feedback.
In contrast to HMO-reported interventions, physicians were asked whether they had received feedback on their STD screening practices from any affiliated Medicaid HMOs, whether they had received STD training in the past 2 years, and whether they had any STD guidelines, including the source of these guidelines (CDC/USPSTF, public health department/state, and other/don't know vs. none). These physician-reported interventions were included because many physicians have multiple Medicaid and commercial HMO contracts and receive a variety of information from these HMOs.
A number of physician personal and practice characteristics were included as covariates and possible alternative explanations for CT screening practices. These included gender, experience (practicing for 10 or fewer years or more than 10 years), specialty (internal medicine, pediatrics, obstetrics/gynecology, or general or family practice), practice type (solo, group, clinic, or other), percent of patients who were Medicaid recipients, physician reimbursement from the medical group/HMO that provided the largest number of Medicaid patients to the practice (capitation, salary, fee-for-service, or unknown), and the number of Medicaid HMOs contracted with by the physician. Finally, we also included the type of CT screening test most frequently used (swab, urine, or other/don't know). Those with unknown responses in several variables were included to avoid loss of sample, but their results were not interpreted.
The extent to which physicians followed the CT screening guideline was analyzed in separate bivariate and multiple regression models. Physician adherence with annual CT screening of sexually active females was assessed separately for patients aged 15 to 19 and 20 to 25 years to include the practices of pediatricians who do not typically see patients more than 19 years of age. The analyses were not weighted because the sampling frame was a simple random sample from the universe of physicians who contracted with the Medicaid HMOs in the selected counties.
The characteristics of physicians in the sample are provided in Table 1. Physicians frequently reported consistently adhering to annual CT screening of young females aged 15 to 19 (58%) and 20 to 25 (68%) years. In comparison, a high percentage of affiliated Medicaid HMOs per physician recommended CT screening of sexually active young females aged 15 to 19 (62%) and 20 to 25 (60%) years. Mailing guidelines and physician training on STD guidelines were reported by 65% of HMOs, however, fewer HMOs per physician incorporated other interventions such as both review and feedback on STD screening (41%) or including quality of care payment criteria in physician contracts (33%). In contrast, 20% of physicians reported having feedback from their affiliated Medicaid HMOs, 47% had received STD training in the past 2 years, and 43% had the CDC or USPSTF CT screening guidelines.
The Relationship of HMO Interventions With CT Screening Rates
Several HMO interventions significantly increased the likelihood of physician adherence with CT screening guidelines in unadjusted bivariate logistic regressions (Table 2). Physicians had significantly higher odds of screening sexually active females aged between 15 and 19 years if a higher percentage of their affiliated Medicaid HMOs recommended that screening (OR = 1.7). They also had higher odds of screening for the following: if a higher percentage of these HMOs trained their contracted physicians on STD guidelines (OR = 1.7); if a higher percentage of these HMOs performed review and feedback of screening practices (OR = 1.8); and if a higher percentage used quality of care as payment criteria (OR = 1.8). The same HMO interventions significantly increased the odds of consistent CT screening rates for females aged 20 to 25 years.
HMO recommendations of screening were highly correlated with HMO training of contracted physicians, providing review and feedback, and using quality of care as payment criteria. The latter 3 variables were excluded from the following adjusted models. The effect of HMO screening recommendation for females aged 15 to 19 was significant (OR = 1.7), but the effect of the other included interventions was no longer significant in models adjusted for physician self-reported interventions and other physician characteristics (Table 3). Frequent CT screening of those aged 15 to 19 years was more likely when physicians reported having received feedback (OR = 1.8), STD training (OR = 1.7), or access to CDC/USPSTF guidelines (OR = 1.9). Frequent CT screening of those aged 20 to 25 years was more likely when the physicians reported STD training in the past 2 years (OR = 1.9).
The odds of consistent CT screening of sexually active females aged 15 to 19 were higher among female physicians (OR = 1.8) and those practicing for 10 of fewer years (OR = 1.8). Odds of screening were lower for pediatricians (OR = 0.3) and internists (OR = 0.5), but higher for those in obstetrics/gynecology (OR = 2.9), relative to family/general practitioners. The exclusion of pediatricians from the models examining screening of 15- to 19-year-olds did not change the results. The odds of consistent screening were also significantly higher for physicians practicing in clinics (OR = 1.7) or other settings (OR = 2.3) compared with solo practitioners. Having fewer Medicaid patients (less than 10% of total patients) significantly reduced the odds of consistent screening (OR = 0.5) compared with physicians with ≥50% Medicaid patients in their practice. Using urine-based CT tests increased the odds of screening (OR = 1.5) compared with swab tests. Similar relationships were found for HMO recommendation to screen females aged 20 to 25 (OR = 1.8), STD training, years of practice, internal medicine specialty, practices with less than 10% Medicaid patients, and use of urine-based tests when examining consistent adherence with annual CT screening of sexually active females aged 20 to 25 years. Each additional Medicaid HMO contract also increased the odds of screening (OR = 1.3).
Our results show that Medicaid HMO interventions such as explicit CT screening recommendations, training of physicians on STD guidelines, review and feedback of physician practices, and incorporating quality of care in payment criteria may increase consistent CT screening among affiliated physicians. Explicit recommendations are important indicators of physician adherence to guidelines and highly correlate with other HMO interventions such as training of physicians, providing review and feedback, and incorporating quality measures as financial incentives. Intervening factors include physician-reported interventions such as training, feedback, and CDC/USPSTF guidelines, as well as the number of individual and practice characteristics of physicians. Notably, the role of intervening factors differed by physician screening given the age of females.
Annual CT screening of sexually active females younger than 26 years of age has been included in CDC guidelines since 1993 and in other organizations' guidelines over the years.31 It was added as a HEDIS measure in 2000—approximately 1½ years before the data collected for this study. The independent association between HMO interventions and physician adherence suggests that HMOs can play a role in changing provider practices with specific interventions and a multipronged approach. For example, HMO interventions such as review and feedback are shown to have mixed effects on physician adherence elsewhere.11,32 Our study indicates that physician acknowledgement of having received feedback from any affiliated Medicaid HMO plays a more important role, at least for screening of those 15 to 19 years of age. In other words, physicians have to specifically acknowledge or perceive such feedback for a noticeable effect on their CT screening patterns.
Similarly, physician-reported recent STD training (for both age groups) and availability of CDC/USPSTF guidelines (for 15–19 year-olds) appear to have the desired effect among physicians. Lack of a significant odds ratio for physician-reported feedback and CDC/USPSTF guidelines for screening of 20 to 25 year olds indicates a differential response to these interventions given the age of patients. It is likely that physician-reported feedback might have not emphasized or included annual screening for the older age groups. However, it is equally likely that physicians did not respond to such feedback about their older patients for clinical reasons or on the basis of past experiences. This explanation may also apply to physician responses to CDC/USPSTF guidelines. In other words, physicians may not follow these guidelines for clinical reasons or depend on their own judgment to determine whether annual screening of sexually active females aged 20 to 25 should be done.
The higher likelihood of CT screening by female physicians is corroborated by other findings.19,33 We find that this effect is only present for younger patients. The higher likelihood of screening among younger/newly graduated physicians is consistent with other findings of an inverse relationship between physician age and adherence to care standards.34 This finding may represent the more negative perceptions of older physicians toward CT screening guidelines that are inconsistent with their original training or less flexibility in changing practice patterns after many years of practice.
The lower likelihood of screening by pediatricians and internists (for both age groups) may reflect physician (mis)perceptions regarding sexual activity among young females in their practice or lower awareness of high CT risk in these age groups.24,33,35–37 It is also likely that internists see fewer young females in the 15 to 25 age group and are therefore less familiar with CT screening guidelines. Obstetrician/gynecologists, however, are more likely to be sensitive to the need for CT screening as may be expected. The lack of a similar effect among patients older than 20 years may be because physicians routinely examine patients for multiple genital/urinary symptoms and do not just rely on a single test for CT.
The higher likelihood of CT screening in clinic or other nonprivate practice settings may indicate the concentration of sexually active female adolescents in such settings, as well as the organizational and clinical efficiency of practice.33 Similarly, the lower likelihood of screening among physicians with fewer Medicaid patients may indicate a lower concentration of high-risk patients. Alternatively, the collective effect of Medicaid HMO contracts seems to increase consistent screening rates, indicating the potential utility of forming coordinated efforts at the organization level or through centralized policy making. The consistent positive effect of urine-based CT testing is most likely because of the ease of this method of screening, which avoids the need for invasive examination. Our findings support the potential utility of this screening method to improve consistent CT screening among all physicians.
This study has limitations and strengths. The findings are from Medicaid HMOs and may not be comparable with fee-for-service Medicaid or providers in other states. However, the factors that motivate physicians to adhere to guidelines are likely to be applicable elsewhere, particularly, because the great majority of physicians who contracted with Medicaid HMOs in this study also practiced in multiple other settings. Also, focusing on Medicaid HMOs has the advantage of identifying the limitations of CT screening among a highly vulnerable and at-risk population and identifying the most effective approaches to curb the spread of CT.
Furthermore, all data were self-reported, and physicians generally report higher rates of compliance in their clinical practices than can be supported by chart reviews.36,37 Although chart reviews are more accurate in identifying physicians' practices, their costs are often more prohibitive than survey methods. Not all constraints and influences on physician adherence, such as the effect of actions by affiliated commercial HMOs, could be examined in this study and additional data are required to examine these complex relationships. More research is also needed to investigate efforts at influencing physician practices such as provision of reminders and prompts, intensive training sessions, or use of targeted financial incentives to promote CT guideline adherence. Although the data used for this study are collected in 2002, the dynamics that have produced these results are still present and are likely to continue to generate similar results in response to guidelines.
The major strength of this study lies in its examination of the relationship of HMO interventions to promote CT screening by physicians independent of a number of important alternative explanations. The paucity of information on HMO level influences on physician practices has been a significant barrier to determine effective methods of improving CT screening rates. The findings indicate that interventions targeting annual CT screening of sexually active young females have to be tailored to characteristics of physicians. In addition, interventions are more likely to be effective if provided in formats that are perceived and acknowledged by physicians.
1.Centers for Disease Control and Prevention. Chlamydia screening among sexually active young female enrollees of health plans–United States, 2000–2007. Morb Mortal Wkly Rep 2009; 58:362–365.
2.Levine WC, Dicker LW, Devine O, et al. Indirect estimation of Chlamydia screening coverage using public health surveillance data. Am J Epidemiol 2004; 160:91–96.
3.Centers for Disease Control and Prevention. Sex transmitted diseases treatment guidelines. Morb Mortal Wkly Rep 2006; 55:1–94.
4.Centers for Disease Control and Prevention. Sexually transmitted diseases treatment guidelines. Morb Mortal Wkly Rep 2002; 51:1–78.
5.Henderson Z, Tao G, Irwin K. Sexually transmitted disease care in managed care organizations. Infect Dis Clin North Am 2005; 19:491–511.
6.Lafferty WE, Kimball AM, Bolan G, et al. Medicaid managed care and STD prevention: Opportunities and risks. J Public Health Manag Pract 1998; 4:52–58.
7.Pourat N, Brown ER, Razack N, et al. Medicaid managed care and STDs: Missed opportunities to control the epidemic. Health Aff (Millwood) 2002; 21:228–239.
8.Christiansen-Lindquist L, Tao G, Hoover K, et al. Chlamydia screening of young sexually active, Medicaid-insured women by race and ethnicity, 2002–2005. Sex Transm Dis 2009; 36:642–646.
11.Flynn KE, Smith MA, Davis MK. From physician to consumer: The effectiveness of strategies to manage health care utilization. Med Care Res Rev 2002; 59:455–481.
12.Leider HL. Influencing physicians: the three critical elements of a successful strategy. Am J Manag Care 1998; 4:583–588; quiz 591–582.
13.Pourat N, Rice T, Tai-Seale M, et al. Association between physician compensation methods and delivery of guideline-concordant STD care: Is there a link? Am J Manag Care 2005; 11:426–432.
14.Shortell SM, Zazzali JL, Burns LR, et al. Implementing evidence-based medicine: The role of market pressures, compensation incentives, and culture in physician organizations. Med Care 2001; 39(7 suppl 1):I62–I78.
15.Strunk BC, Reschovsky JD. Kinder and Gentler: Physicians and Managed Care, 1997–2001. Washington, DC: Center for Studying Health System Change, 2002. Tracking Report No. 5.
16.Bindman AB, Wholey DR, Christianson JB. Physicians' reports of their experience with health plan care management practices. Am J Manag Care 2003; 9(spec no 2):SP100–SP110.
17.Reed M, Devers K, Landon B. Physicians and care management: More acceptance than you think. Issue Brief Cent Stud Health Syst Change 2003; 60:1–4.
18.Cabana MD, Rand CS, Powe NR, et al. Why don't physicians follow clinical practice guidelines? A framework for improvement. JAMA 1999; 282:1458–1465.
19.Wiesenfeld HC, Dennard-Hall K, Cook RL, et al. Knowledge about sexually transmitted diseases in women among primary care physicians. Sex Transm Dis 2005; 32:649–653.
20.Flores G, Lee M, Bauchner H, et al. Pediatricians' attitudes, beliefs, and practices regarding clinical practice guidelines: A national survey. Pediatrics 2000; 105(3 pt 1):496–501.
21.Christianson JB, Wholey DR, Warrick L, et al. How are health plans supporting physician practice? The physician perspective. Health Aff (Millwood) 2003; 22:181–189.
22.Chorba T, Scholes D, Bluespruce J, et al. Sexually transmitted diseases and managed care: An inquiry and review of issues affecting service delivery. Am J Med Qual 2004; 19:145–156.
23.Ray MN, Wall T, Casebeer L, et al. Chlamydia screening of at-risk young women in managed health care: Characteristics of top-performing primary care offices. Sex Transm Dis 2005; 32:382–386.
24.Torkko KC, Gershman K, Crane LA, et al. Testing for Chlamydia and sexual history taking in adolescent females: Results from a statewide survey of Colorado primary care providers. Pediatrics 2000; 106:E32.
25.Magid DJ, Stiffman M, Anderson LA, et al. Adherence to CDC STD guideline recommendations for the treatment of Chlamydia trachomatis infection in two managed care organizations. Sex Transm Dis 2003; 30:30–32.
26.Lafferty WE, Downey L, Holan CM, et al. Provision of sexual health services to adolescent enrollees in Medicaid managed care. Am J Public Health 2002; 92:1779–1783.
27.Keil L, Chattopadhayay M, Potter F, et al. Community Tracking Study Physician Survey: Survey Methodology Report (Round 1, Release 1). Rockville, MD: The Gallup Organization, 1988. Technical Publication No. 9.
29.Pourat N, Nihalani J, Razack N, et al. Sexually Transmitted Disease Services in California's Medi-Cal Managed Care: Findings From a Baseline Survey of STD Care Delivery in 2002. Los Angeles, CA: UCLA Center for Health Policy Research, 2006.
31.Walsh C, Anderson LA, Irwin K. The silent epidemic of Chlamydia trachomatis: The urgent need for detection and treatment in women. J Womens Health Gend Based Med 2000; 9:339–343.
32.Hillman AL, Ripley K, Goldfarb N, et al. The use of physician financial incentives and feedback to improve pediatric preventive care in Medicaid managed care. Pediatrics 1999; 104(4 pt 1):931–935.
33.Cook RL, Wiesenfeld HC, Ashton MR, et al. Barriers to screening sexually active adolescent women for Chlamydia: A survey of primary care physicians. J Adolesc Health 2001; 28:204–210.
34.Choudhry NK, Fletcher RH, Soumerai SB. Systematic review: The relationship between clinical experience and quality of health care. Ann Intern Med 2005; 142:260–273.
35.McDougall L, Mathias RG, O'Connor BA, et al. Management of Chlamydia trachomatis genital infections: Reported practices of primary care physicians. CMAJ 1992; 146:715–721.
36.Garbutt J, Jeffe DB, Shackelford P. Diagnosis and treatment of acute otitis media: An assessment. Pediatrics 2003; 112(1 pt 1):143–149.
37.Maue SK, Segal R, Kimberlin CL, et al. Predicting physician guideline compliance: An assessment of motivators and perceived barriers. Am J Manag Care 2004; 10:383–391.