Effectiveness of Enhanced Primary Care on Preventive Health Services

, , , , ,
The American Journal of Accountable Care, September 2018, Volume 6, Issue 3

This natural experiment compared rates of indicated preventive care for low-income Hispanic patients enrolled in an enhanced primary care program with those of patients receiving usual care.

ABSTRACT

Objectives: Team-based primary care programs have the potential to address healthcare disparities. We evaluated the impact of a multidisciplinary team-based enhanced primary care program on vaccination and cancer screening rates at a community health center serving a low-income Hispanic community.

Study Design: Quasi-experimental.

Methods: In this controlled before-and-after study, data were extracted from the electronic health record for the following groups of patients aged 18 to 64 years: (1) those enrolled in the program between August 2011 and January 2012 and (2) randomly selected controls cared for during this time. Outcomes included eligible patients’ receipt of (1) influenza and/or pneumococcal vaccines and (2) breast, cervical, and/or colon cancer screens over a 12-month period following enrollment. We used an adjusted difference-in-differences (DID) analysis to estimate the effect of the intervention.

Results: A total of 134 intervention and 393 control patients were included in the analysis. Unadjusted influenza vaccination rates increased from 47.3% to 63.6% in the control group and from 48.5% to 63.4% in the intervention group in adjusted models, a nonsignificant DID of —1.4 percentage points (95% CI, –14.1 to 8.5; P = .66). Pneumococcal vaccination rates fell for the control (63.7% to 57.4%) and intervention (53.9% to 42.6%) groups, with a nonsignificant DID of —4.7 percentage points (95% CI, –16.5 to 7.0; P = .45). There was no difference in changes in cancer screening rates.

Conclusions: A newly implemented culturally and linguistically concordant enhanced primary care program had no measurable effect on preventive care outcomes for a low-income, largely Hispanic population. These findings suggest a need for a deeper examination of both the implementation process of similar programs and their long-term effectiveness.

Am J Accountable Care. 2018;9(3):9-15Many team-based systems of care, such as case management programs, peer health coaching, and patient-centered medical home certification, seek to enhance the delivery of primary care services.1-9 Healthcare organizations and insurance companies have both played a role in developing and funding these types of programs in an effort to improve primary care services and reduce costs.10 The term “enhanced primary care” refers to a team-based care model that aims to improve care processes and patient outcomes for both preventive care services and chronic disease management.11,12

Enhanced primary care programs are often directed at populations at highest risk for underutilization of preventive services, overutilization of acute care services, and health disparities. Disparities in preventive care services related to race/ethnicity, language, and socioeconomic status contribute to a disproportionate risk of disease among vulnerable populations in the United States.13-18 An enhanced primary care team may improve access to care through increased personalized care and may develop more trusting relationships, potentially benefitting patients who experience disparities in care the most. In this study, we aimed to determine the impact of a newly implemented, insurance company—sponsored, team-based enhanced primary care program (Buena Salud) on vaccination and cancer screening rates in a low-income, largely Hispanic population with high rates of limited English proficiency.

METHODS

Design, Setting, and Participants

Buena Salud is a bilingual enhanced primary care program for Medicaid managed care patients in western Massachusetts. The program was financed by the Health New England insurance company and implemented by Brightwood Health Center (BHC). BHC is an urban community health center that serves a predominantly Hispanic population (88%) who are insured primarily by either Medicaid (59%) or Medicare (28%). More than 50% of BHC patients prefer Spanish as their spoken language. Buena Salud was started in an effort to improve compliance with preventive care, provide support for chronic disease management, and optimize utilization of healthcare services. Most patients were referred to Buena Salud by their primary care provider, but patients also were enrolled through a periodic auto-enrollment process or self-referral. The Buena Salud team consisted of 2 registered nurse care managers, 2 medical assistants trained as community health workers, and a case worker. The care team differed from many other managed care teams in that each team member was bilingual (Spanish/English) and was from the same racial/ethnic group as the majority of BHC patients. The Buena Salud team offered education and coaching for disease self-management, provided additional support for patients in the clinic and at home, used electronic health registries to identify patients in need of care and services such as vaccination and cancer screenings, and provided linkage to community-based support services. A more detailed description of the program is included in the eAppendix (available at ajmc.com). Each Buena Salud nurse was expected to actively manage up to 50 patients at any given time. Care intensity varied depending on individual patients’ needs.19

We reviewed eligible patients’ electronic health records (EHRs), extracting data pertinent to the study aims. Eligibility for the intervention arm of the study was defined as all BHC patients aged 18 to 64 years who (1) were newly enrolled in the Buena Salud program between August 1, 2011, and January 31, 2012 (intervention period); (2) were eligible for at least 1 of the outcomes measured; and (3) remained a patient at BHC for 12 months following enrollment in Buena Salud. Eligibility criteria for controls included all BHC patients aged 18 to 64 years who (1) had been seen for a clinical encounter at BHC in the intervention period, (2) were not enrolled in Buena Salud, and (3) remained a patient at BHC for 12 months following enrollment in the study. From the list of eligible controls, we randomly selected 3 patients who had been seen in the clinic in the same month that an intervention patient had been enrolled per 1 intervention patient (3:1 matching) using a random number generator. We selected controls based on when they were seen in the clinic rather than matching on variables such as age, race/ethnicity, or gender because (1) the population attending the clinic is relatively homogeneous for these measurable variables and (2) selecting patients seen in a similar time frame reduces the potential for differences in unmeasured confounders related to temporal changes in practice at the health center. Intervention patients and controls must have had at least 1 visit at BHC in the 12 months prior to the enrollment/index visit so that baseline data could be extracted. The methodology is similar to that used in a study of the impact of the Buena Salud program on outcomes for patients with diabetes.19 This study was approved by the Baystate Institutional Review Board, which waived informed consent.

Time Period Studied and Outcome Variables

The primary study outcome was the change in the percentage of eligible patients who received recommended vaccinations and cancer screenings prior to implementation of the Buena Salud program to 12 months post implementation. Eligibility for influenza and pneumococcal vaccinations and breast, cervical, and colon cancer screenings was based on national guidelines from the following organizations: (1) CDC for vaccines,20,21 (2) United States Preventive Services Task Force (USPSTF) guidelines for breast and colon cancer screening,22,23 and (3) American College of Obstetrics and Gynecology guidelines for cervical cancer screening.24 Patients were considered up to date on vaccination and/or cancer screenings at baseline as follows: (1) influenza vaccination in the 12 months prior to enrollment, (2) pneumococcal vaccine consistent with CDC guidelines depending on patient’s comorbidities and age, (3) mammogram to screen for breast cancer in the 2 years prior to enrollment for patients 50 years or older, (4) Pap smear to screen for cervical cancer in the 3 years prior to enrollment for patients 21 years or older, and (5) any of the colon cancer screening modalities consistent with USPSTF guidelines for patients 50 years or older. We chose the time period following enrollment (12 months) so that we could include the maximum number of intervention patients in the study. In addition to vaccination and cancer screening data, we also extracted data on potential confounders (eg, comorbidities and the number of years a patient received care at BHC) from participants’ EHRs (Table 1).

Data Extraction

We first oriented data extractors to the study’s data dictionary and extraction protocol, which included where to locate pertinent data in the EHR. After establishing baseline consistency using standardized extraction forms, 2 extractors independently reviewed 20 randomly selected health records to assess interextractor consistency at 6 and 18 months into the course of data extraction to test whether consistency was maintained.19 Study data were collected and managed using REDCap electronic data capture tools.25

Analysis

Participant characteristics are presented as means and SDs for continuous variables and frequencies and percentages for categorical variables. To estimate differences between the groups studied for vaccinations, we used a difference-in-differences (DID) approach (the difference between the pre-post change in the intervention group compared with the control group). Outcomes were modeled using generalized estimating equations with exchangeable correlations and robust standard errors (clustering on patient). Outcomes were modeled using the logit link and binomial family. Models were estimated with main effects for the intervention group and time period, with an interaction term between these 2 representing the DID. Because participants in the intervention group were frequency matched to controls based on enrollment month, we used enrollment month as an indicator variable in the model. This term was not significant in the models, so it was therefore removed. Statistical significance was set at an α of .05. Multivariable models included potential confounders (eg, demographic data, comorbidities, and number of years receiving care at BHC). Using Wald tests, models were reduced to include variables that were significant at the .05 level. To control for possible residual confounding and for face validity, we retained age, mental health, and substance use in all models. The analysis was conducted using Stata version 13.1 (StataCorp LP; College Station, Texas).

RESULTS

A total of 527 patients were included in the study: 134 in the intervention group and 393 in the control group. The median age was 50 years (range, 20-65), 72.7% were female, and Spanish was the preferred language for 56.7% (Table 1). Most of those in the intervention group (estimated 90%) were referred to the program by a healthcare provider, with fewer (estimated 10%) enrolled via auto-enrollment or self-referral. Baseline differences between the groups included the following: The control group had a slightly higher median age compared with the intervention group (50 vs 49 years), were more likely to have HIV (10.2% vs 1.5%), were more likely to have been prescribed buprenorphine or naloxone (6.6% vs 2.2%), were less likely to be homeless (1.0% vs 2.2%), and were less likely to have been diagnosed with anxiety or depression (25.7.8% vs 29.9%) (Table 1). The intervention group had a slightly higher number of visits with a primary care provider (0.7 vs 1.5 visits; DID, 0.8; P = .03) in the 12 months following enrollment. All other variables were similar at baseline. Extractor agreement was estimated at 93%.

Vaccination Rates

Influenza. All patients were eligible for influenza vaccination. Overall, influenza vaccination rates increased from 47.6% in the year prior to enrollment to 63.6% at the end of the study period. The increase was similar for controls (47.3% to 63.6%) and intervention patients (48.5% to 63.4%). The adjusted difference in the change in percentage points (—2.8; 95% CI, –14.1 to 8.5; P = .66) between groups was not significant (Table 2).

Pneumococcal. Of the 527 patients included in the study, 432 were eligible for pneumococcal vaccination (317 control and 115 intervention). Overall, the percentage of eligible patients who were up to date on pneumococcal vaccination decreased from 61.6% prior to inception of the Buena Salud program to 53.5% 15 months following the start of the program. The unadjusted vaccination rate for controls went from 63.7% to 53.4% and from 53.9% to 42.6% for the intervention group. The adjusted difference in change between the groups was not significant, at —4.7 percentage points (95% CI, –16.5 to 7.0; P = .45) (Table 2).

Cancer Screening Rates

Breast cancer screening. Of the 527 patients included in the study, 178 were eligible to be screened for breast cancer (131 control and 47 intervention). Of these, 147 (82.6%) were up to date on breast cancer screening at the time of enrollment (baseline) and 82.0% were up to date at the end of the study period. The control group’s unadjusted screening rates went from 86.3% to 85.5% and the intervention group’s rate did not change from 72.3%, resulting in an adjusted difference of —0.7 percentage points (95% CI, –9.2 to 10.6; P = .83) (Table 2).

Cervical cancer screening. Of the 527 patients included in the study, 153 were eligible for cervical cancer screening (113 control and 40 intervention). Of these, 72.6% were up to date on screening at the time of enrollment and 79.6% were up to date at the end of the study period. Control patients’ unadjusted screening rates increased from 72.6% to 81.4% and intervention arm rates went from 72.5% to 75.0%, giving a nonsignificant adjusted difference in rate changes between groups of —6.3 percentage points (95% CI, –22.5 to 9.9; P = .40) (Table 2).

Colon cancer screening. Of the 527 patients included in the study, 273 were eligible for colon cancer screening (202 control and 71 intervention). Overall, 50.9% of patients in the study were up to date on colon cancer screening at baseline and 52.8% were up to date at the end of the study period. Unadjusted screening rates increased from 54.0% to 56.4% in the control group and did not change from 42.3% in the intervention group, for a DID of —2.5 percentage points (95% CI, –14.8 to 9.8; P = .70) (Table 2).

DISCUSSION

In this controlled before-and-after study, we found that patients enrolled in an ethnically and culturally concordant team-based enhanced primary care program implemented in a community health center serving a population of lower-income, largely Hispanic patients did not experience significantly different changes in rates of vaccination or cancer screenings compared with patients not enrolled in the program. The study demonstrates some of the complexities of programs designed to enhance care delivery processes in underserved communities and raises questions about the most appropriate outcomes to measure in the early stages of interventions that are intended to improve patient outcomes and decrease costs in a racially and ethnically diverse population.

Racial and ethnic disparities in vaccination rates and cancer screenings persist in the US healthcare system.13,14,16,17,26,27 Explanations for these disparities are likely multifactorial, including system-, provider-, and patient-level factors. Prior interventions to reduce disparities in cancer screening have taken both population-and clinic-based approaches. Interventions that have demonstrated effectiveness include a cluster-randomized trial of 150,417 patients in the United Kingdom. This study found a reduction in the socioeconomic status gradient in screening with an enhanced reminder letter.28 A systematic review of patient navigators also showed improved screening rates for breast, cervical, and colon cancer for patients with limited English proficiency,29 and tracking and outreach at the clinic level have been effective for increasing influenza vaccination in an inner-city population.30 Enhanced primary care teams, particularly teams staffed by a racially/ethnically concordant staff, have the potential to reduce disparities in preventive care services through outreach and by addressing issues related to limited English proficiency, mistrust in the healthcare system, and implicit bias, similar to patient navigator interventions. Our study’s results do not demonstrate differences between intervention and control groups for the outcomes measured in the current study. Rather than conclude that the program is ineffective, we reflected on whether the outcomes that were measured, which are common measures of enhanced primary care programs’ effectiveness, were the best measures of intervention effectiveness in the context of a new program implemented in a natural setting. The following provides some insights into these reflections.

Interventions to improve both vaccination and cancer screening rates have shown success in experimental settings, but the enhanced primary care model we studied was implemented outside of a clinical trial and there may be a learning curve for the enhanced care team.31 We followed patients for a relatively brief period (15 months) after enrollment in the program, and it may take more time for team members to develop trusting relationships with care recipients, thus limiting the early effects on these “hard” outcomes. We learned through interviews with the Buena Salud team that there was no systematic process for documenting the team’s interactions with patients during the time period studied. This meant that we could not accurately measure the intervention “dose” individuals received, meaning some patients enrolled may have had no “touch” by the team. In a small study such as this, variation in expertise among the Buena Salud team members also could have influenced the outcomes observed.19

Strengths and Limitations

This study’s strengths included the following: comparison with a control group, use of a DID analysis that adjusted for secular trends in care and outcomes, and risk adjustment using a broad array of clinical and demographic data. The latter allowed us to address the nonrandom assignment of patients to intervention and control groups. This study should also be considered in light of its limitations. First, this was an observational study and not a randomized clinical trial. The Buena Salud program sought to provide support for the sickest patients, as evidenced by the measured baseline differences found between Buena Salud patients and controls; there may have been other unmeasured important differences not accounted for in our risk adjustment. Second, this study evaluated patients from 1 health center. Although this allowed us to focus on the population of interest, the intervention might have different effects in other populations or in other health centers with a similar population. Third, we were unable to assess whether patients were offered vaccination or cancer screenings but declined. Fourth, screening guidelines vary somewhat, and it is possible that application of guidelines other than those used in this study could result in different estimates of uptake. Finally, enrollment in the Buena Salud program was associated with a slightly greater number of office visits with a primary care provider; these additional visits may have presented an additional burden for the intervention group.

CONCLUSIONS

A team-based enhanced primary care program delivered by a multidisciplinary bilingual team that was linguistically and culturally concordant with the majority of the underserved patients enrolled in the program did not have statistically significant effects on vaccination and cancer screening rates in its first 15 months of existence. Care should be taken in drawing conclusions from the outcomes assessed in the first year of a new program because there is likely a learning curve to engaging and partnering with patients in this context. Additional outcomes, such as measures of patient experience, trust, and satisfaction with care, may yield more meaningful insight into a program’s effectiveness. Studies that examine the implementation and longitudinal effectiveness of these programs will contribute additional important information to our understanding of the potential benefits of enhanced primary care team interventions for at-risk patient populations.Author Affiliations: University of Massachusetts Medical School—Baystate (SLG, AK, HG-K, AG, PKL), Springfield, MA; Western New England Renal and Transplant Associates (LM), Springfield, MA.

Source of Funding: This study was funded by an incubator grant from Baystate Medical Center. Dr Goff was supported by the National Institute for Child Health and Human Development of the National Institutes of Health under award number K23HD080870. Dr Lindenauer was supported by grant K24HL132008 from the National Heart, Lung, and Blood Institute.

Author Disclosures: Dr Goff, Ms Murphy, Mr Knee, Ms Guhn-Knight, Dr Guhn, and Dr Lindenauer are or were employed by Baystate Health, which owns Health New England, a health insurance company that financially supports the Be Healthy program.

Authorship Information: Concept and design (SLG, LM, AG, PKL); acquisition of data (SLG, LM, HG-K, AG); analysis and interpretation of data (SLG, AK, PKL); drafting of the manuscript (SLG, AK, HG-K); critical revision of the manuscript for important intellectual content (SLG, LM, AK, HG-K, AG, PKL); statistical analysis (SLG, AK); provision of study materials or patients (AG); obtaining funding (SLG); administrative, technical, or logistic support (HG-K); and supervision (SLG, PKL).

Send Correspondence to: Sarah Goff, MD, University of Massachusetts Medical School—Baystate, 759 Chestnut St, Springfield, MA 01199. Email: Sarah.goffmd@baystatehealth.org.REFERENCES

1. Brown SA, García AA, Winter M, Silva L, Brown A, Hanis CL. Integrating education, group support, and case management for diabetic Hispanics. Ethn Dis. 2011;21(1):20-26.

2. Comino EJ, Davies GP, Krastev Y, et al. A systematic review of interventions to enhance access to best practice primary health care for chronic disease management, prevention and episodic care. BMC Health Serv Res. 2012;12:415. doi: 10.1186/1472-6963-12-415.

3. Edwards ST, Bitton A, Hong J, Landon BE. Patient-centered medical home initiatives expanded in 2009-13: providers, patients, and payment incentives increased. Health Aff (Millwood). 2014;33(10):1823-1831. doi: 10.1377/hlthaff.2014.0351.

4. Ghorob A, Bodenheimer T. Sharing the care to improve access to primary care. N Engl J Med. 2012;366(21):1955-1957. doi: 10.1056/NEJMp1202775.

5. Goldman ML, Ghorob A, Hessler D, Yamamoto R, Thom DH, Bodenheimer T. Are low-income peer health coaches able to master and utilize evidence-based health coaching? Ann Fam Med. 2015;13(suppl 1):S36-S41. doi: 10.1370/afm.1756.

6. Landon BE. Moving ahead with the PCMH: some progress, but more testing needed. J Gen Intern Med. 2013;28(6):753-755. doi: 10.1007/s11606-013-2434-0.

7. Shojania KG, Ranji SR, McDonald KM, et al. Effects of quality improvement strategies for type 2 diabetes on glycemic control: a meta-regression analysis. JAMA. 2006;296(4):427-440. doi: 10.1001/jama.296.4.427.

8. Thom DH, Ghorob A, Hessler D, De Vore D, Chen E, Bodenheimer TA. Impact of peer health coaching on glycemic control in low-income patients with diabetes: a randomized controlled trial. Ann Fam Med. 2013;11(2):137-144. doi: 10.1370/afm.1443.

9. Thom DH, Hessler D, Willard-Grace R, et al. Health coaching by medical assistants improves patients’ chronic care experience. Am J Manag Care. 2015;21(10):685-691.

10. Gibson TB, Wang S, Kelly E, et al. A value-based insurance design program at a large company boosted medication adherence for employees with chronic illnesses. Health Aff (Millwood). 2011;30(1):109-117. doi: 10.1377/hlthaff.2010.0510.

11. Sperl-Hillen J, O’Connor PJ, Carlson RR, et al. Improving diabetes care in a large health care system: an enhanced primary care approach. Jt Comm J Qual Improv. 2000;26(11):615-622.

12. Ozieh MN, Bishu KG, Dismuke CE, Egede LE. Trends in health care expenditure in U.S. adults with diabetes: 2002-2011. Diabetes Care. 2015;38(10):1844-1851. doi: 10.2337/dc15-0369.

13. Coughlin SS, Blumenthal DS, Seay SJ, Smith SA. Toward the elimination of colorectal cancer disparities among African Americans. J Racial Ethn Health Disparities. 2016;3(4):555-564. doi: 10.1007/s40615-015-0174-z.

14. Miller JW, Plescia M, Ekwueme DU. Public health national approach to reducing breast and cervical cancer disparities. Cancer. 2014;120(suppl 1):2537-2539. doi: 10.1002/cncr.28818.

15. Joseph DA, Redwood D, DeGroff A, Butler EL. Use of evidence-based interventions to address disparities in colorectal cancer screening. MMWR Suppl. 2016;65(1):21-28. doi: 10.15585/mmwr.su6501a5.

16. Bleser WK, Miranda PY, Jean-Jacques M. Racial/ethnic disparities in influenza vaccination of chronically ill US adults: the mediating role of perceived discrimination in health care. Med Care. 2016;54(6):570-577. doi: 10.1097/MLR.0000000000000544.

17. Lu P, O’Halloran A, Williams WW, Lindley MC, Farrall S, Bridges CB. Racial and ethnic disparities in vaccination coverage among adult populations in the U.S. Vaccine. 2015;33(suppl 4):D83-D91. doi: 10.1016/j.vaccine.2015.09.031.

18. Williams WW, Lu P-J, O’Halloran A, et al; CDC. Surveillance of vaccination coverage among adult populations - United States, 2014. MMWR Surveill Summ. 2016;65(1):1-36. doi: 10.15585/mmwr.ss6501a1.

19. Goff SL, Murphy L, Knee AB, Guhn-Knight H, Guhn A, Lindenauer PK. Effects of an enhanced primary care program on diabetes outcomes. Am J Manag Care. 2017;23(3):e75-e71.

20. Influenza ACIP vaccine recommendations. CDC website. cdc.gov/vaccines/hcp/acip-recs/vacc-specific/flu.html. Updated July 6, 2018. Accessed July 12, 2016.

21. Vaccine-specific recommendations (by date published). CDC website. cdc.gov/vaccines/hcp/acip-recs/recs-by-date.html. Updated July 6, 2018. Accessed July 12, 2016.

22. Final update summary: breast cancer: screening. US Preventive Services Task Force website. uspreventiveservicestaskforce.org/Page/Document/UpdateSummaryFinal/breast-cancer-screening. Updated January 2016. Accessed July 12, 2016.

23. Final recommendation statement: colorectal cancer: screening. US Preventive Services Task Force website. uspreventiveservicestaskforce.org/Page/Document/RecommendationStatementFinal/colorectal-cancer-screening. Updated October 2014. Accessed July 12, 2016.

24. Armstrong C. Practice guidelines: ACOG updates guidelines on cervical cytology screening. American Family Physician website. www.aafp.org/afp/2010/0601/p1380.html. Published June 2010. Accessed July 12, 2016.

25. Harris PA, Taylor R, Thielke R, Payne J, Gonzalez N, Conde JG. Research electronic data capture (REDCap)—a metadata-driven methodology and workflow process for providing translational research informatics support. J Biomed Inform. 2009;42(2):377-381. doi: 10.1016/j.jbi.2008.08.010.

26. Almario CV, May FP, Maxwell AE, Ren W, Ponce NA, Spiegel BMR. Persistent racial and ethnic disparities in flu vaccination coverage: results from a population-based study. Am J Infect Control. 2016;44(9):1004-1009. doi: 10.1016/j.ajic.2016.03.064.

27. Brown T, Goldman SN, Acosta F, et al. Understanding black patients’ refusal of pneumococcal vaccination. J Racial Ethn Health Disparities. 2017;4(1):1-8. doi: 10.1007/s40615-015-0193-9.

28. Wardle J, von Wagner C, Kralj-Hans I, et al. Effects of evidence-based strategies to reduce the socioeconomic gradient of uptake in the English NHS Bowel Cancer Screening Programme (ASCEND): four cluster-randomised controlled trials. Lancet. 2016;387(10020):751-759. doi: 10.1016/S0140-6736(15)01154-X.

29. Genoff MC, Zaballa A, Gany F, et al. Navigating language barriers: a systematic review of patient navigators’ impact on cancer screening for limited English proficient patients. J Gen Intern Med. 2016;31(4):426-434. doi: 10.1007/s11606-015-3572-3.

30. Humiston SG, Bennett NM, Long C, et al. Increasing inner-city adult influenza vaccination rates: a randomized controlled trial. Public Health Rep. 2011;126(suppl 2):39-47. doi: 10.1177/00333549111260S206.

31. Optimize the care team. Institute for Healthcare Improvement website. ihi.org/resources/Pages/Changes/OptimizetheCareTeam.aspx. Accessed January 21, 2016.