Advertisement

Journal of General Internal Medicine

, Volume 33, Issue 11, pp 1845–1847 | Cite as

Cost Conversations Between Primary Care Providers and Patients with Expanded Medicaid Coverage

  • Renuka TipirneniEmail author
  • Minal R. Patel
  • Matthias A. Kirch
  • Susan D. Goold
Concise Research Reports

KEY WORDS

out-of-pocket costs health insurance medicaid physician-patient communication 

INTRODUCTION

Patients face increasing out-of-pocket (OOP) costs for healthcare,1 which have been associated with medication non-adherence and poor health outcomes.2, 3 While low-income patients may frequently have concerns about OOP costs—even if they have insurance with generous covered benefits—they may not raise cost concerns with physicians. Little is known about cost conversations between primary care providers (PCPs) and low-income patients. Our objective was to determine the frequency, predictors, and PCPs’ perceptions of the impact of cost conversations with low-income patients in an expanded Medicaid program in Michigan (“Healthy Michigan Plan” [HMP]), a state program for adults ages 19–64 with incomes ≤ 138% of the federal poverty level (FPL) and includes limited cost-sharing for beneficiaries (≤ 2% of income).

METHODS

We conducted a mailed survey of all PCPs caring for ≥ 12 HMP patients in June–November 2015. The sample was derived from the Michigan Department of Health and Human Services (MDHHS) Medicaid claims data warehouse and included both physician and non-physician (nurse practitioner or physician assistant) PCPs. The University of Michigan and MDHHS institutional review boards considered the study exempt.

Respondents were asked, “Have you ever discussed out-of-pocket medical costs with a HMP patient?” Respondents who answered “yes” were asked: (a) “Thinking of the most recent time you discussed out-of-pocket medical expenses with a HMP patient, who brought up the topic?” and (b) “Thinking of the most recent time you discussed out-of-pocket medical expenses with a HMP patient, did the conversation result in a change in the management plan for the patient?”

Descriptive statistics report responses to individual survey items. χ2 and logistic regression analyses were used to examine the unadjusted and adjusted associations between PCPs’ personal, professional, and practice characteristics, and (1) the likelihood of cost conversations, and (2) change in management due to cost conversations. Statistical analyses were performed using Stata version 14.2; two-sided p < 0.05 was considered significant.

RESULTS

The response rate was 56% (N = 2104). Respondent PCPs were 55% male, 79% White, 5% Black, 2% Hispanic/Latino, and 83% physicians. Four hundred forty-five (22%) said that they had discussed OOP medical costs with a HMP patient. Of those who had cost conversations, the topic was brought up 56% of the time by the patient, 39% by the PCP, 4% by somebody else in the practice (e.g., clerical or nursing staff), and 1% by another person. When a cost conversation occurred, 56% of PCPs reported it resulted in a change in management.

Cost conversations were more frequent among female, White, and non-physician PCPs, those with prior care for the underserved, working in federally qualified health centers, or in practices with Medicaid/uninsured-predominant payer mixes or rural settings (Table 1). Changes in management due to cost conversations were more common among non-physician PCPs, those with fewer years in practice and in rural settings. In multivariable regression analyses, the adjusted odds of cost conversations were greater for White, Hispanic/Latino, and non-physician PCPs, and those with Medicaid/uninsured-predominant payer mixes (Table 2). PCPs with fewer years in practice or in non-suburban settings had greater adjusted odds of management changes due to cost conversations.
Table 1

Association of PCP Personal, Professional, and Practice Characteristics with Frequency of Cost Conversations and Change in Clinical Management Due To Cost Conversations

PCP characteristics

All respondents

N (col %)

Cost conversations

n (row %§)

Change in management due to cost conversation

n (row %†)

Total

2104 (100%)

445/1988 (22.4%)

248/440 (56.4%)

Personal characteristics

 Gender

  Male

1165 (55.4%)

227/1107 (20.5%)§

118/224 (52.7%)

  Female

939 (44.6%)

218/881 (24.7%)

130/216 (60.2%)

 Race‡

  White

1579 (79.1%)

367/1145 (24.3%)

204/364 (56.0%)

  Black/African American

92 (4.6%)

14/91 (15.4%)

8/14 (57.1%)

  Asian/Pacific Islander

218 (10.9%)

25/204 (12.3%)

14/23 (60.9%)

  Other/more than one

107 (5.4%)

18/103 (17.5%)

10/18 (55.6%)

 Ethnicity‡

  Hispanic/Latino

46 (2.3%)

15/45 (33.3%)

8/15 (53.3%)

  Not Hispanic/Latino

1978 (97.7%)

416/1475 (22.0%)

234/411 (56.9%)

Professional characteristics

 Provider type

  Physician

1750 (83.2%)

337/1653 (20.4%)

180/333 (54.1%)

  Non-physician (NP or PA)

357 (16.8%)

108/335 (32.2%)

68/107 (63.6%)

 Specialty

  Family medicine

1123 (53.4%)

230/1064 (21.6%)

119/228 (52.2%)§

  Internal medicine

574 (27.3%)

96/538 (17.8%)

58/94 (61.7%)

  Other physician specialty

53 (2.5%)

11/51 (21.6%)

3/11 (27.3%)

  Non-physician (NP or PA)

354 (16.8%)

108/335 (32.2%)

68/107 (63.6%)

 Years in practice‡

  < 10 years

520 (25.9%)

126/502 (25.1%)

87/125 (69.6%)§

  10–20 years

676 (33.7%)

134/644 (20.8%)

72/133 (54.1%)

  > 20 years

810 (40.4%)

172/753 (22.8%)

84/169 (49.7%)

 Prior care for underserved patients‡

  Yes

1153 (57.0%)

284/1102 (25.8%)

161/282 (57.1%)

  No

871 (43.0%)

151/834 (18.1%)

82/148 (55.4%)

Practice characteristics

 Practice size‡

  Small (≤ 5 providers)

1157 (57.5%)

252/1087 (23.2%)

141/250 (56.4%)

  Large (> 5 providers)

855 (42.5%)

181/820 (22.1%)

103/178 (57.9%)

 FQHC practice‡

  Yes

311 (14.9%)

94/299 (31.4%)

58/94 (61.7%)

  No

1770 (85.1%)

347/1669 (20.8%)

188/343 (54.8%)

 University/teaching hospital practice‡

  Yes

276 (13.4%)

48/263 (18.3%)

27/47 (57.5%)

  No

1786 (86.6%)

388/1687 (23.0%)

217/384 (56.5%)

 Hospital-based practice (non-teaching)‡

  Yes

643 (31.2%)

134/609 (22.0%)

82/132 (62.1%)

  No

1419 (68.8%)

302/1341 (22.5%)

162/299 (54.2%)

 Payer mix‡

  Medicaid/uninsured predominant

689 (36.0%)

177/670 (26.4%)§

104/177 (58.8%)

  Private/Medicare/other predominant

1223 (64.0%)

232/1160 (20.0%)

128/230 (55.7%)

 Urbanicity

  Urban

1584 (75.3%)

312/1492 (20.9%)§

168/309 (54.4%)§

  Suburban

193 (9.2%)

42/185 (22.7%)

20/42 (47.6%)

  Rural

327 (15.5%)

91/311 (29.3%)

60/89 (67.4%)

*Row percent among respondents who answered the question about cost conversations (N = 1988)

†Row percent among those respondents who had a cost conversation (N = 440)

‡All respondents column does not sum to 2104 due to skipped responses

§p < 0.05

p < 0.001

Table 2

Multivariable Association of PCP Personal, Professional, and Practice Characteristics with Likelihood of Cost Conversations, and Likelihood of Change in Clinical Management Due To Cost Conversations

PCP characteristics

Adjusted odds ratio*

(95% CI)

Odds of cost conversation

Odds of change in management due to cost conversation†

Personal characteristics

 Male gender

0.82 (0.63–1.05)

0.91 (0.58–1.41)

 Race

  White

[ref]

[ref]

  Black/African American

0.52 (0.28–0.96)‡

0.92 (0.29–2.93)

  Asian/Pacific Islander

0.43 (0.27–0.70)‡

1.37 (0.54–3.46)

  Other/More than one

0.65 (0.36–1.17)

1.60 (0.52–4.94)

Ethnicity, Hispanic/Latino

2.11 (1.08–4.12)‡

0.93 (0.31–2.77)

Professional characteristics

 Provider type, physician (ref = non-physician)

0.71 (0.51–0.99)‡

0.96 (0.54–1.73)

 Years in practice

  < 10 years

[ref]

[ref]

  10–20 years

0.81 (0.60–1.09)

0.52 (0.30–0.89)‡

  > 20 years

1.04 (0.77–1.42)

0.47 (0.27–0.82)‡

Practice characteristics

 Payer mix

  Medicaid/uninsured predominant

1.31 (1.02–1.69)‡

0.95 (0.60–1.51)

  Private/Medicare/other predominant

[ref]

[ref]

 Urbanicity

  Urban

0.82 (0.60–1.11)

0.62 (0.35–1.11)

  Suburban

0.70 (0.45–1.11)

0.41 (0.18–0.95)‡

  Rural

[ref]

[ref]

*Each column represents a different multivariable model

†Odds of change in management among those respondents who had a cost conversation

p < 0.05

§p < 0.001

DISCUSSION

Only one in five PCPs reported conversations about out-of-pocket medical costs with low-income Medicaid patients. The frequency of cost conversations we observed appears on the low-end range of 4–65% observed in other studies examining cost conversations in the general population.4

However, half of PCPs who had cost conversations reported a resulting change in management, attesting to the value of such conversations for patient-centered care, even for Medicaid patients with generous covered benefits and limited cost-sharing. Although OOP costs may be declining for some groups after Affordable Care Act coverage expansion, even nominal OOP costs may be considerably burdensome for low-income patients.2, 3 Thus, greater investment is needed to improve the frequency and quality of cost conversations.5, 6

Potential study limitations include self-reported outcomes, survey questions limited to Medicaid expansion patients in a single state, a sample enriched for PCPs caring for at least 12 Medicaid expansion patients, the study’s focus on assessing cost conversations from the PCP perspective, and the absence of data on types of changes in management. Frequency of cost conversations may differ for other healthcare provider or patient groups, or in other states. Future research may examine facilitators and barriers to cost conversations, and resulting management changes.

Notes

ACKNOWLEDGMENTS

We gratefully acknowledge the contribution of other evaluation team members who contributed to survey development and administration including Eric G. Campbell, PhD; John Z. Ayanian, MD MPP; Cengiz Salman, MA; Sarah J. Clark, MPH; Tammy Chang, MD MPH MS; Adrianne N. Haggins, MD MSc; Edith C. Kieffer, PhD; Lisa Szymecko, JD PhD; Sunghee Lee, PhD; Erica Solway, PhD MPH MSW; Erin Beathard, MPH MSW; and Zachary Rowe, BS.

FUNDING INFORMATION

The study was funded by a contract from the Michigan Department of Health and Human Services (MDHHS) to the University of Michigan to conduct an evaluation of the Healthy Michigan Plan, as required by the Centers for Medicare and Medicaid Services (CMS) through a Section 1115 Medicaid waiver.

Disclaimer

The views expressed in this article are those of the authors and do not necessarily reflect the position or policy of MDHHS or CMS.

COMPLIANCE WITH ETHICAL STANDARDS

Conflict of Interest

The authors declare that they do not have a conflict of interest.

REFERENCES

  1. 1.
    Paez KA, Zhao L, Hwang W. Rising out-of-pocket spending for chronic conditions: a ten-year trend. Health Aff. 2009;28(1):15–25.CrossRefGoogle Scholar
  2. 2.
    Patel MR, Piette JD, Resnicow K, Kowalski-Dobson T, Heisler M. Social Determinants of Health, Cost-related Nonadherence, and Cost-reducing Behaviors Among Adults With Diabetes: Findings From the National Health Interview Survey. Med Care. 2016;54(8):796–803.CrossRefGoogle Scholar
  3. 3.
    Heisler M, Choi H, Rosen AB, et al. Hospitalizations and deaths among adults with cardiovascular disease who underuse medications because of cost: a longitudinal analysis. Med Care. 2010;48(2):87–94.CrossRefGoogle Scholar
  4. 4.
    Hunter WG, Hesson A, Davis JK, et al. Patient-physician discussions about costs: definitions and impact on cost conversation incidence estimates. BMC Health Serv Res. 2016;16:108.CrossRefGoogle Scholar
  5. 5.
    Alexander GC, Casalino LP, Meltzer DO. Patient-physician communication about out-of-pocket costs. JAMA. 2003;290(7):953–958.CrossRefGoogle Scholar
  6. 6.
    Piette JD, Heisler M, Wagner TH. Cost-related medication underuse: do patients with chronic illnesses tell their doctors? Arch Intern Med. 2004;164(16):1749–1755.CrossRefGoogle Scholar

Copyright information

© Society of General Internal Medicine 2018

Authors and Affiliations

  • Renuka Tipirneni
    • 1
    • 2
    Email author
  • Minal R. Patel
    • 1
    • 3
  • Matthias A. Kirch
    • 1
  • Susan D. Goold
    • 1
    • 2
    • 3
    • 4
  1. 1.Institute for Healthcare Policy and InnovationUniversity of MichiganAnn ArborUSA
  2. 2.Division of General Medicine, Department of Internal MedicineUniversity of MichiganAnn ArborUSA
  3. 3.School of Public HealthUniversity of MichiganAnn ArborUSA
  4. 4.Center for Bioethics and Social Sciences in MedicineUniversity of MichiganAnn ArborUSA

Personalised recommendations