Trends in Self-payment for Outpatient Psychiatrist Visits

Abstract

This study uses survey data to assess changes in the rates of self-payment for outpatient psychiatric care compared with primary care.

Even though insurance coverage for mental health has greatly improved over the last 10 years in the US,^1,2 many patients continue to struggle to find psychiatrists willing to accept their insurance and need to pay upfront for their psychiatrist visits.³ This is a hurdle that many patients cannot surmount, even if a portion of that payment is eventually paid by insurance. This study aimed to explore patterns in self-payment for office-based psychiatric services and changes over time, particularly with the passage of the Mental Health Parity and Addiction Equity Act and the Affordable Care Act.^4,5

Methods

This study used 10 years of data (January 2007 through December 2016) from the National Ambulatory Medical Care Survey (NAMCS),⁶ which is a random sample of outpatient visits made to a nationally representative random sample of physicians who are nonfederally employed, younger than 85 years, and treating outpatients. Patients seen in hospital units, nursing homes or other extended care institutions, or the patient's home were not included. This dataset also does not include telephone visits. Of note, prior to 2012, the NAMCS included community health centers in their primary sampling frame. Starting in 2012, these were sampled separately. As a result, they are excluded from this analysis.

The study accounted for complex survey design to describe the characteristics of patients who self-paid (submitting either self-payment or upfront payment that the patient later submitted to their insurance for at least partial reimbursement) and clinicians who were reimbursed predominantly (>75%) by self-payment for office-based visits. We also compared payment trends for psychiatrist visits with payment trends for primary care clinician visits using logistic regression with group by time interaction. The NAMCS has a waiver of consent from the National Center for Health Statistics’ Research Ethics Review Board because it used secondary data and posed minimal risk to participants. Institutional review board approval was waived because the NAMCS data set is publicly available. Data analysis was completed from December 2019 to February 2020 with StataIC version 15 (StataCorp), with a significance threshold set at a 2-tailed P < .05.

Results

After we excluded visits to community health centers, there were 16 464 psychiatrist visits in the sample (mean [SE], 20.1 [13.7] visits per psychiatrist), of which 15 790 had expected source of payment information. There were 127 500 primary care visits in the sample (mean [SE], 26.4 [13.4] visits per primary care clinician), of which 119 749 had expected source of payment information. There were 816 psychiatrists in the sample and 750 with information on predominant source of revenue. There were 4842 primary care physicians in the sample and 4294 with information on the predominant source of revenue.

Of the psychiatrist visits, 3445 (21.8%; weighted, 22.0%) were self-paid by patients, compared with 4336 primary care clinician visits (3.6%; weighted, 3.6%). One hundred forty-six psychiatrists (19.5%; weighted, 23.5%) were reimbursed predominantly by self-payment, compared with 69 primary care clinicians (1.6%; weighted, 1.7%).

As shown in the Figure, the percentage of visits to psychiatrists that patients self-paid has trended upward (from 18.5% in 2007-2009 to 26.7% in 2014-2016), while the percentage of visits to primary care clinicians that patients self-paid has trended downward (from 4.1% in 2007-2009 to 2.8% in 2014-2016). The percentage of psychiatrists who work in predominantly self-pay practices has trended upward (from 16.4% in 2007-2009 to 26.4% in 2014-2016), while the percentage of primary care clinicians who work in predominantly self-pay practices has not changed significantly (from 1.5% to 1.7%).

Figure. Graphical Display and Logistic Regression With Group by Time Interaction for Longitudinal Trends in Self-payment for Office-Based Visits.

The period 2007-2009 is prior to implementation of the Paul Wellstone and Pete Domenici Mental Health Parity and Addiction Equity Act (MHPAEA), 2010-2013 is the period after implementation of MHPAEA but prior to implementation of the Affordable Care Act (ACA), and 2014-2016 is the period after implementation of many provisions of the ACA. The MHPAEA “prevents group health plans and health insurance issuers that provide mental health or substance use disorder (MH/SUD) benefits from imposing less favorable benefit limitations on those benefits than on medical/surgical benefits.”⁴ The ACA requires “that most individual and small employer health insurance plans, including all plans offered through the Health Insurance Marketplace, cover mental health and substance use disorder services.”⁵ Pursuant to ACA implementation, 15% of Americans had no insurance in 2013, vs 9% in 2015.² If psychiatrists accepted insurance, increased insurance coverage should reduce the percentage of patients who self-pay and the percentage of psychiatrists predominantly paid by patients directly for office-based psychiatric care. Visit-level analysis uses visit-level weights and is based on response of yes to the statement “expected source of payment: self” by the clinician. Physician-level analysis uses clinician-level weights. Metrics are based on a response of “more than 75%” to the question “Roughly what percent of your patient care revenue comes from patient payments?”

At the visit level, we found that self-payment for psychiatrist visits was significantly more common among white patients (white patients, 3082 of 12 732 [24.2%; weighted, 24.2%]; black patients, 97 of 1191 [8.1%; weighted, 9.7%]; Hispanic patients, 173 of 1193 [14.5%; weighted, 12.2%]; P < .001) and male patients (1587 of 6952 men [22.8%; weighted, 24.5%]; 1858 of 8838 women [21.0%; weighted, 20.2%]; P < .001) and not significantly different across age groups (<18 years, 353 of 2318 individuals [15.2%; weighted, 16.8%]; 18-64 years, 2765 of 11 832 individuals [23.4%; weighted, 23.5%]; >64 years, 327 of 1640 individuals [19.9%; weighted, 19.3%]; P = .08). Self-paid visits were a mean (SE) of 38.3 (1.1) minutes in duration, as opposed to 28.8 (0.7) minutes for visits paid directly by third parties (P < .001). Patients who were self-paying had a mean (SE) of 18.3 (2.1) visits in the 12 months prior to the current visit compared with 9.4 (0.6) visits for patients with third-party payers (P < .001).

We found that psychiatrists who are reimbursed predominantly by self-payment were more likely to work in solo practices (mean [SE], 30.5% [2.8%]) than group practices (8.3% [2.3%]; P < .001) and were less likely to care predominantly for pediatric patients (mean [SE], 6.6% [4.3%]) than adult patients (25.4% [2.3%]; P = .01). Compared with those receiving fewer self-payments, psychiatrists reimbursed predominantly by self-payments saw fewer total office-based outpatients per week (mean [SE], 21.0 [0.8] visits vs 15.0 [1.2] visits; P < .001), had a greater mean percentage of white patients (mean [SE], 77.6% [0.1%] vs 87.3% [1.8%]; P < .001), had longer mean appointment times (mean [SE] minutes, 31.6 [0.7] vs 40.5 [1.6]; P < .001), and saw patients more frequently (mean [SE] visits per patient in the last 12 months, 10.3 [0.7] vs 20.4 [2.2]; P < .001) (Table).

Table. Characteristics of Psychiatrists Whose Office-Based Practices Are Predominantly Reimbursed by Self-payment (>75%).

Categories	Psychiatrists predominantly reimbursed by self-payment, % (SE)	P value
Categorical variables
Practice type
Solo practice	30.5 (2.8)	<.001
Partnership or group practice	8.3 (2.3)
Practice ownership
Physician or physician group	29.4 (2.6)	<.001
Medical/academic health center or other hospital	5.1 (5.0)
Insurance company, health plan, health maintenance organization, or other health care corporation	4.3 (3.1)
Population focus
Pediatric (>50% of patients <18 y)	6.6 (4.3)	.01
Adult	25.4 (2.3)
Region
Northeast	30.6 (4.3)	.07
Midwest	13.7 (4.0)
South	21.0 (3.2)
West	23.6 (4.8)
Continuous variables, by self-payment reimbursement levela,b,c
Mean office-based visits/wk, No.
≤75%	21.0 (0.8)	<.001
>75%	15.8 (1.2)
Mean appointment time, min
≤75%	31.6 (0.7)	<.001
>75%	40.5 (1.6)
Mean visits/patient in last 12 mo
≤75%	10.3 (0.7)	<.001
>75%	20.4 (2.2)
White patients, % (SE)
≤75%	77.6 (0.1)	<.001
>75%	87.3 (1.8)

^aUniverse includes office-based psychiatrists not employed by the federal government or community health centers. The response rate ranged from 24.0%-53.2%.

^bCharacteristics of psychiatrists in the sample: 265 of 750 (64.7%; weighted, 68.5%) were in solo practice; 535 (71.3%; weighted, 75.0%) in practices that are physician or physician-group owned, 36 (4.8%; weighted, 4.8%) in practices owned by hospitals or medical centers, and 116 (15.5%; weighted, 13.5%) for health plans or other corporations; 458 (61.1%; weighted, 64.9%) were full owners, 110 (14.7%; weighted, 15.2%) part owners, 132 (17.6%; weighted, 15.0%) employees, and 49 (6.5%; weighted, 4.8%) contractors; 72 (9.6%; weighted, 10.3%) had a pediatric focus (>50% of patients <18 years); 204 psychiatrists (27.2%; weighted, 30.5%) were located in the Northeast, 137 (18.3%; weighted, 14.9%) in the Midwest, 224 (29.9%; weighted, 28.8%) in the South, and 185 (24.7%; weighted, 25.7%) in the West; 164 individuals (21.9%; weighted, 26.2%) were in the sample from 2007-2009, 395 (52.7%; weighted, 38.5%) from 2010-2013, and 191 (25.5%; weighted, 35.5%) from 2014-2016.

^cThis is a physician-level analysis based on a response of “more than 75%” to the question “Roughly what percent of your patient care revenue comes from patient payments?”

Conclusions

Despite the small sample of psychiatrists in the NAMCS and the associational design of this study, this study appears to find that many patients continue to self-pay for psychiatrist visits and many psychiatrists continue to only care for patients that can self-pay. Psychiatrists may be more likely to rely on self-payment models than other specialties because of low insurance reimbursement rates, particularly for psychotherapy, as well as a demand for psychiatric services that outstrips supply. Our findings begin to highlight a 2-tiered system for outpatient psychiatrist care, which presents potential issues of health equity. Patients who choose to or must rely on third-party payment for psychiatrist appointments may be receiving only psychopharmacology, while those who can self-pay may be able to receive psychotherapy as well.