Unmet healthcare needs among midlife adults with mental distress and multiple chronic conditions

Abstract

Background:

Limited attention has focused on midlife health. Yet, this is a time of great change, including onset of chronic conditions and changes in mental health.

Objective:

To examine unmet healthcare needs among midlife adults (50–64 years) in the US with severe psychological distress (SPD) and/or multiple chronic conditions (MCC).

Methods:

Nationally representative data for midlife adults (50–64 years) from NHIS 2014–2018 were examined (n = 39,329). Multimorbidity status: no MCC/SPD, MCC alone, SPD alone, or both. We used logistic regression to estimate adjusted odds ratios (AOR) of delayed or foregone care by multimorbidity status.

Results:

Nearly 40% of midlife adults had MCC, SPD, or SPD/MCC. SPD with or without MCC had higher prevalence of social disadvantage, fair/poor health, activity limitations, and delayed/foregone healthcare. Compared to those with neither, adults with SPD/MCC were more likely to delay care due to limited office hours (AOR = 4.2, 95% CI 3.1–5.5) and had nearly three to four times higher odds of delays for all other reasons. Those with SPD/MCC had higher odds of needing but not getting mental healthcare (AOR = 6.4, 95% CI 4.5–9.1), prescriptions (AOR = 4.8, 95% CI 3.9–5.9), or follow-up care (AOR = 5.0, 95% CI 3.7–6.6), and three to four times higher odds of all other types of foregone care.

Conclusions:

Midlife adults with SPD/MCC have substantial unmet healthcare needs. Midlife is a critical time to manage both chronic conditions and mental illness. Coordinated efforts by policy-makers and healthcare systems are crucial to address complex healthcare needs of this population at a critical stage of the life-course.

Introduction

Premature mortality in those with serious mental illness is a public health crisis. Adults with mental illness are more likely to have multiple chronic conditions (MCC) and die, on average, 10–25 years earlier than the general population, primarily due to cardiovascular disease and other chronic conditions (Roberts et al., 2017; Walker, McGee, & Druss, 2015). Survey data suggest that around 25% of US adults currently have a mental illness, and 50% will experience a mental illness at some time in their lives (Centers for Disease Control & Prevention, 2011). Mental illness is the second largest contributor to years lived with disability, with mental and behavioral disorders accounting for, on average, 27 years (Murray, Atkinson, & Bhalla, 2013) and $24 billion in annual disability benefits (Centers for Disease Control & Prevention, 2011). In addition to the direct costs of mental healthcare, mental illness exacerbates chronic conditions and chronic disease morbidity, leading to substantial indirect costs (Centers for Disease Control & Prevention, 2011; Insel, 2008). Combined, this has resulted in 90% of the $3.3 trillion in annual US healthcare expenditures spent on treating those with chronic physical and mental health conditions (Buttorff, Ruder, & Bauman, 2017).

Access to healthcare during midlife (ages 50 to 64) is critical for prevention and management of chronic health conditions, as nearly half of midlife adults have two or more (Smolka, Purvis, & Figueiredo, 2009). Limited attention has been focused on midlife health (Lachman, 2015). Yet, this is a time of great change, including onset of functional limitations, disability, chronic conditions, and changes in mental health (Harlow & Derby, 2015). Lifestyle, social interaction, and preventive care behaviors can have important impacts on health in older age. Exercise and staying fit in midlife is associated with better cognitive function in later life (Lachman, Teshale, & Agrigoroaei, 2015), and healthy behaviors in midlife are shown to predict healthy aging (Sabia et al., 2012). Adopting a healthy lifestyle, even in midlife, has demonstrated health benefits (King, Mainous, & Geesey, 2007). Intervening in midlife can enhance the transition into healthy later life, and healthcare providers have an important role.

Much attention has been paid to healthcare delivery models that address the complex needs of patients with chronic conditions (Bodenheimer, Wagner, & Grumbach, 2002; Coleman, Austin, Brach, & Wagner, 2009; Stellefson, Dipnarine, & Stopka, 2013). However, the emphasis of these models has primarily been physical health. Until recently, mental health and behavioral healthcare in the US have been dissociated from physical health and healthcare, although the two are inextricably linked. With the passage of the Mental Health Parity and Addiction Equity Act (MHPAEA) in 2008, there was an explicit acknowledgement that mental health treatment should be on par with physical health treatment and they should be covered comparably by large-group, employer-based health insurers. The Patient Protection and Affordable Care Act (PPACA) of 2010 further mandated coverage of mental health and substance use services by small-group, individual, and Medicaid expansion plans (“Patient Protection and Affordable Care Act,” 2010).

Although coverage for mental healthcare has improved, it remains siloed from general medical care, and access continues to be a challenge. Of particular concern is the lack of coordinated care for those with both chronic physical and mental health conditions. Those with serious mental illness are at increased risk of cardiometabolic disease due both to genetic and pathophysiological differences associated with mental illness and to the effects of psychotropic medications (Penninx & Lange, 2018). Thus, it is imperative for patients with mental illness to have metabolic risks monitored and managed and have chronic condition care take into account the unique needs of patients with mental illness. Studies have examined the impact of integrated or co-located mental healthcare and primary care for managing multimorbid patients (Kilbourne, Lai, Bowersox, Pirraglia, & Bauer, 2011; Pirraglia, Kilbourne, Rosen, & Witt, 2011). However, little is known about the magnitude of the problem in the general midlife population and potential unmet healthcare needs among those with severe psychological distress (SPD) and/or MCC. Understanding relationships between multimorbid mental and chronic physical conditions and access to healthcare services in midlife adults is critical to identifying and addressing barriers to care during this important period in the lifecourse.

The purpose of this paper is to address this gap by examining unmet healthcare needs in a nationally representative sample of US midlife adults who have SPD, MCC, or both. The aims of this study are to: 1) document differences in characteristics of this population, 2) identify differences in health services access and utilization, and 3) examine the odds of delayed and forgone healthcare among midlife adults with MCC alone, SPD alone, and both SPD and MCC, compared to neither.

Methods and materials

Data source and target population

Data were from a nationally representative sample of midlife adults, ages 50–64 (n = 40,430 unweighted), who participated in the 2014–2018 National Health Interview Survey (NHIS). The NHIS is an annual household survey about the health and healthcare of the US non-institutionalized, civilian population that uses a multistage probability sample designed to be representative of the US population (Gentleman & Pleis, 2002; National Center for Health Statistics, 2017). Our analytic sample included midlife adults who had complete data for all key variables (n = 39,329 unweighted; 97.3% of the total). Data were de-identified and publicly available, which does not meet requirements for human subjects’ research, so the study was exempt from review.

Measures

Severe psychological distress (SPD) was defined using the K6 population screening tool, which quantifies psychological and somatic symptoms that are common to many mental disorders but not specific to any one disorder (McVeigh et al., 2006). The K6 comprises six questions asking respondents how often they experienced symptoms in the past 30 days (i.e. felt sad, nervous, hopeless, restless, worthless, or that everything was an effort) (Kessler et al., 2002). Response options were 1-All of the time, 2-Most of the time, 3-Some of the time, 4-A little of the time, and 5-None of the time. Responses were reverse coded from 0 to 4 and summed to yield a K6 score ranging from 0 to 24, with higher scores indicating greater severity of psychological distress (Kessler et al., 2003; Pirraglia, Hampton, Lai, Friedmann, & O’Toole, 2011). We used standard cut-points of 13–24 to indicate severe psychological distress (SPD) (Kessler et al., 2003; Prochaska, Sung, Max, Shi, & Ong, 2012). Multiple chronic conditions (MCC) were defined as two or more self-reports of having been told by a healthcare professional that one had: diabetes, hypertension, coronary heart disease, stroke, cancer, arthritis, hepatitis, kidney disease, asthma, or chronic obstructive pulmonary disease (Ward, Schiller, & Goodman, 2014). Multimorbidity status was the cross-classification of SPD and MCC, which yielded four categories indicating the presence or absence of each: MCC alone, SPD alone, both MCC and SPD, and neither.

Healthcare outcomes

We examined two types of healthcare outcomes: delayed care and foregone care. Delayed care includes indicators representing reasons for delaying needed care in the past year: cost, unable to get appointment, limited office hours, too long in the waiting room, unable to get through on the phone, or transportation issues. Foregone care includes indicators of types of care or services needed but not received due to cost: medical, medical specialist, follow-up, mental health, or dental care, prescription medications, or eyeglasses.

Covariates

Following Andersen’s Behavioral Model of Health Services Use, covariates represented predisposing factors, enabling factors, and need factors (Andersen, 1995; Andersen & Aday, 1978). Predisposing factors included: sex, age, race/ethnicity, educational attainment, and marital status. Enabling factors included employment, poverty (below 200% of the federal poverty level vs. above), insurance status, and having a usual source of care (USOC). Need factors included self-reported health and limitations to activities of daily living.

Analysis

First, we examined differences in predisposing and enabling factors, health status, and health services use by multimorbidity status using cross-tabulations with design-based F-tests. Next, we used logistic regression models to examine differences in reasons for delayed care and types of foregone care by multimorbidity status adjusted for covariates listed above. Sampling weights were adjusted to account for five years of pooled data. Analyses were conducted with Stata SE version 15 and accounted for unequal probability of selection and the complex sample design of NHIS (“Stata Statistical Software,” 2017).

Results

Population characteristics

Nearly 40% of midlife adults had either MCC alone (35.6%), SPD alone (1.2%), or SPD and MCC combined (2.9%). Table 1 shows predisposing, enabling, and need factors of the study population and differences by multimorbidity status for each characteristic. For predisposing factors, those with SPD with or without MCC were more likely to be female and less likely to be college educated or married compared to those with neither SPD nor MCC. With respect to enabling factors, those with SPD and MCC were substantially less likely to be employed (18%) compared to those with SPD alone (37%), MCC alone (55%), or neither (76%). Among those with any SPD or MCC, the majority of those not employed were due to health or disability: SPD/MCC (61%), SPD alone (39%), MCC alone (23%). Those with SPD alone were the least likely to be insured or have a usual source of care. For need factors, overall, 20% of midlife adults reported a limitation to their activities of daily living. However, 77% of those with SPD/MCC reported any activity limitation compared to 56% of those with SPD alone, 34% of those with MCC alone, and 9% of those with neither.

Table 1.

Characteristics of midlife adults (50–64 years) by multimorbidity status, 2014–2018.

	No SPD or MCC	MCC only	SPD only	SPD and MCC	Total	P-value
Predisposing factors
Sex
Female	51.0%	51.5%	61.4%	60.0%	51.6%	<0.001
Male	49.0%	48.5%	38.6%	40.0%	48.4%
Age group
50–54 years	40.0%	26.7%	42.3%	33.2%	35.1%	<0.001
55–59 years	33.1%	33.7%	38.2%	34.9%	33.4%
60–64 years	26.9%	39.5%	19.6%	31.9%	31.5%
Race/Ethnicity
White	71.0%	69.9%	69.8%	68.4%	70.5%	<0.001
Black	10.0%	14.7%	11.7%	12.9%	11.8%
AIAN	0.5%	1.1%	1.1%	1.4%	0.7%
Asian	6.0%	3.8%	2.7%	1.5%	5.0%
Hispanic	12.5%	10.6%	14.8%	15.9%	12.0%
Educational attainment
Less than a H.S. diploma	9.3%	13.5%	18.9%	26.0%	11.4%	<0.001
High school diploma	23.9%	28.0%	32.0%	34.2%	25.7%
Some college	15.8%	18.6%	15.6%	19.3%	16.9%
College degree	51.0%	39.9%	33.5%	20.5%	46.0%
Marital status
Married	67.9%	60.9%	48.0%	41.8%	64.4%	<0.001
Separated, divorced, widowed	19.4%	24.7%	32.9%	40.0%	22.0%
Never married	8.3%	9.7%	12.7%	10.9%	8.9%
Living with partner	4.4%	4.8%	6.4%	7.4%	4.7%
Enabling factors
Employment status
Employed	76.0%	55.5%	37.0%	17.8%	66.5%	<0.001
Unemployed	3.2%	3.7%	6.4%	7.5%	3.5%
Not working, other	7.3%	5.5%	10.6%	7.0%	6.7%
Retired	8.4%	12.1%	7.4%	6.7%	9.7%
Not working, health or disability	5.1%	23.2%	38.6%	61.1%	13.6%
Poverty status
At or above 200% FPL	81.4%	69.9%	47.6%	34.6%	75.5%	<0.001
Below 200% FPL	18.6%	30.1%	52.4%	65.4%	24.5%
Insurance coverage
Insured	90.0%	93.5%	78.1%	88.4%	91.0%	<0.001
Uninsured	10.1%	6.5%	21.9%	11.6%	9.0%
Insurance typesa
Private	78.7%	66.4%	43.4%	31.4%	72.5%	<0.001
Medicaid	6.0%	14.6%	19.7%	35.9%	10.1%	<0.001
Medicare	3.2%	14.8%	17.1%	34.3%	8.4%	<0.001
Usual source of care
Has a usual source	89.4%	95.9%	85.9%	93.9%	91.8%	<0.001
No usual source	10.6%	4.1%	14.1%	6.1%	8.2%
Need factors
Self-reported health status
Excellent/very good/good	93.0%	69.6%	53.5%	26.6%	82.2%	<0.001
Fair/poor	7.0%	30.4%	46.5%	73.5%	17.8%
Any activity limitation
No limitation	91.5%	66.0%	44.5%	23.0%	79.8%	<0.001
Limited in any way	8.5%	34.1%	55.5%	77.0%	20.2%
Unweighted sample size	23,008	14,412	557	1,352	39,329
Weighted population	36,930,744	21,803,868	757,715	1,794,595	61,286,922

^aCategories not mutually exclusive.

Healthcare access

Table 2 shows significant differences across all measures of healthcare utilization, delayed care, and foregone care by multimorbidity status. Overall, 76% of midlife adults saw a general doctor in the past year and about one-third saw a medical specialist, with those having any MCC more likely to have done so. Around 8% of all midlife adults saw a mental health professional in the past year, while 39% of those with SPD/MCC and 30% of those with SPD alone did. In the past year, those with SPD/MCC were significantly more likely to have reported overnight hospitalizations (26% vs. 4%) and to have had two or more ER visits (31% vs. 3%) compared to those with neither, respectively.

Table 2.

Healthcare access among midlife adults (50–64 years) by multimorbidity status, 2014–2018.

	No SPD or MCC	MCC only	SPD only	SPD and MCC	Total	P-value
Care received in past year
Saw general doctor	69.8%	86.8%	71.2%	86.5%	76.4%	<0.001
Saw medical specialist	23.9%	47.2%	38.7%	54.0%	33.3%	<0.001
Saw mental health professional	5.2%	10.1%	29.9%	39.4%	8.3%	<0.001
Hospital overnight, past year	4.2%	14.7%	13.0%	25.7%	8.7%	<0.001
ER visits, past year
None	88.5%	74.0%	64.6%	49.9%	81.9%	<0.001
One visit	8.6%	15.3%	19.5%	18.8%	11.5%
Two or more visits	2.9%	10.7%	15.9%	31.3%	6.7%
Delayed care in past year a
Due to cost	7.5%	13.1%	31.2%	30.6%	10.5%	<0.001
Could not get appointment soon enough	5.0%	9.5%	14.2%	21.9%	7.2%	<0.001
Wait too long in doctor’s office	3.4%	6.0%	11.4%	17.6%	4.8%	<0.001
Could not get through on phone	1.9%	3.8%	5.6%	11.3%	2.9%	<0.001
Office not open when you could go	2.1%	3.8%	8.0%	10.9%	3.0%	<0.001
No transportation	0.8%	3.3%	9.6%	13.2%	2.2%	<0.001
Number of reasons for delayed care
No delays	85.1%	74.0%	53.3%	46.7%	79.7%	<0.001
One delay	11.0%	17.8%	29.4%	27.2%	14.1%
Two or more delays	3.9%	8.3%	17.3%	26.1%	6.3%
Needed but did not get due to cost in past year a
Dental care	8.6%	15.6%	39.3%	43.1%	12.5%	<0.001
Eyeglasses	5.3%	10.5%	29.6%	35.8%	8.3%	<0.001
Prescription medicine	3.9%	11.2%	24.6%	34.3%	7.7%	<0.001
Medical care	5.1%	9.6%	27.9%	29.2%	7.7%	<0.001
Medical specialist	3.1%	7.0%	21.8%	23.9%	5.3%	<0.001
Follow up care	2.4%	5.2%	18.6%	21.3%	4.2%	<0.001
Mental health care	1.0%	2.1%	16.2%	15.7%	2.0%	<0.001
Number of types of foregone care
No foregone care	86.0%	73.8%	45.0%	36.5%	79.7%	<0.001
One type of care foregone	7.1%	11.5%	13.3%	18.6%	9.1%
Two or more types of care foregone	6.9%	14.7%	41.7%	45.0%	11.2%

^aCategories not mutually exclusive.

Midlife adults with SPD or with SPD/MCC were more likely to have a hospitalization in the past year than those with neither (AOR 2.4, 95% CI 1.9–2.9 and AOR= 2.2, 95% CI 2.0–2.5), respectively. Similarly, compared to those with neither SPD nor MCC, those with SPD and those with both SPD and MCC were more likely to have had an ER visit in the past year (AOR 2.7, 95% CI 2.1–3.3 and AOR= 2.2, 95% CI 2.0–2.5), respectively. Midlife adults with MCC alone were not significantly different that those with neither. [Data not shown]

The number of reasons reported for delaying needed care in the past year was higher among those with any SPD compared to those without, with 26% of SPD/MCC and 17% of SPD alone reporting two or more reasons compared to 3% of those with neither. Similarly, the number of types of care needed but foregone due to cost in the past year was higher among those with any SPD compared to those without, with 45% of SPD/MCC and 42% of SPD alone reporting two or more types of care foregone compared to 15% of those with MCC alone and 7% of those with neither.

Delayed care

Table 3 shows that compared to those with neither, adults with SPD/MCC had over four times higher odds of delaying care due to limited office hours (AOR = 4.2, 95% CI 3.1–5.5) and over three times higher odds of delayed care due to difficulty getting through on the phone (AOR = 3.4, 95% CI 2.5–4.8) or inability to get an appointment soon enough (AOR = 3.2, 95% CI 2.6–4.0). They had between 2.6 and 3.4 times higher odds for all other reasons. Those with SPD alone were most likely to have delayed care due to transportation issues (AOR = 3.1, 95% CI 2.1–4.6) or due to limited office hours (AOR = 3.1, 95% CI 1.9–5.0) compared to those with neither. Those with MCC alone were also significantly more likely to report delayed care compared with those who had neither, with between 1.5 and 1.7 times higher odds for each of the reasons.

Table 3.

Adjusted odds ratios of delayed care in past year for midlife adults (50–64 years) by multimorbidity status, NHIS 2014–2018.

	No SPD or MCC	MCC only	SPD only	SPD and MCC
Delayed care due to cost, past 12 months
Adjusted odds ratio	1.0	1.6	2.8	2.6
95% confidence intervals		(1.5, 1.8)	(2.1, 3.7)	(2.1, 3.2)
Delayed care because could not get through by phone, past 12 months
Adjusted odds ratio	1.0	1.6	1.9	3.4
95% confidence intervals		(1.3, 1.9)	(1.1, 3.1)	(2.5, 4.8)
Delayed care because could not get appointment soon enough
Adjusted odds ratio	1.0	1.7	2.2	3.2
95% confidence intervals		(1.5, 1.9)	(1.6, 3.0)	(2.6, 4.0)
Delayed care because wait too long in doctor’s office, past 12 months
Adjusted odds ratio	1.0	1.5	2.4	3.4
95% confidence intervals		(1.3, 1.8)	(1.7, 3.6)	(2.6, 4.4)
Delayed care because doctor’s office not open, past 12 months
Adjusted odds ratio	1.0	1.7	3.1	4.2
95% confidence intervals		(1.4, 2.0)	(1.9, 5.0)	(3.1, 5.5)
Delayed care because lacked transportation, past 12 months
Adjusted odds ratio	1.0	1.7	3.1	3.0
95% confidence intervals		(1.4, 2.1)	(2.1, 4.6)	(2.2, 4.1)

Note. All models adjusted for sex, age, race/ethnicity, education, marital status, employment status, poverty, insurance status, USOC, poor health status, and activity limitations.

Foregone care due to cost

Table 4 shows that those with SPD/MCC had higher odds of needing but not getting mental healthcare (AOR = 6.4, 95% CI 4.5–9.1), prescriptions (AOR = 4.8, 95% CI 3.9–5.9), and follow-up care (AOR = 4.8, 95% CI 3.6–6.4), and had three to four times higher odds of all other types of foregone care due to cost. Those with SPD alone had higher odds of foregone mental healthcare (AOR = 7.3), follow-up care (AOR = 3.9), specialty care (AOR = 3.7), and prescriptions (AOR = 3.3) compared to those with neither. Similarly, those with MCC alone were significantly more likely to report foregone care compared with those who had neither, with between 1.4 and 2.4 times higher odds for each type of care foregone.

Table 4.

Adjusted odds ratios of foregone care due to cost for midlife adults (50–64 years) by multimorbidity status, NHIS 2014–2018.

	No SPD or MCC	MCC only	SPD only	SPD and MCC
Needed but could not afford medical care, past 12 months
Adjusted odds ratio	1.0	1.7	3.3	3.2
95% confidence intervals		(1.5, 1.9)	(2.5, 4.3)	(2.6, 4.0)
Needed but could not afford follow-up care, past 12 months
Adjusted odds ratio	1.0	1.9	3.9	4.8
95% confidence intervals		(1.6, 2.3)	(2.5, 6.1)	(3.6, 6.4)
Needed but could not afford to see a specialist, past 12 months
Adjusted odds ratio	1.0	1.9	3.7	4.0
95% confidence intervals		(1.6, 2.3)	(2.6, 5.2)	(3.1, 5.3)
Needed but could not afford mental health care, past 12 months
Adjusted odds ratio	1.0	1.6	7.3	6.4
95% confidence intervals		(1.2, 2.0)	(5.1, 10.6)	(4.5, 9.1)
Needed but could not afford prescription medicines, past 12 months
Adjusted odds ratio	1.0	2.4	3.3	4.8
95% confidence intervals		(2.1, 2.7)	(2.4, 4.4)	(3.9, 5.9)
Needed but could not afford dental care, past 12 months
Adjusted odds ratio	1.0	1.4	1.3	1.6
95% confidence intervals		(1.3, 1.6)	(2.3, 3.9)	(2.4, 3.6)
Needed but could not afford eyeglasses, past 12 months
Adjusted odds ratio	1.0	1.6	3.2	3.5
95% confidence intervals		(1.4, 1.8)	(2.4, 4.2)	(2.9, 4.3)

Note. All models adjusted for sex, age, race/ethnicity, education, marital status, employment status, poverty, insurance status, USOC, poor health status, and activity limitations.

Discussion

This study fills a gap in the literature by demonstrating differences in access to care among those with mental distress and chronic disease in a representative sample of US midlife adults. While 39% of midlife adults had multiple chronic conditions overall and 4% had severe psychological distress overall, those with any SPD or MCC had substantially more unmet healthcare needs compared to those who had neither. Moreover, midlife adults with SPD (with or without MCC) had poorer physical health and more limitations in activities of daily living due to health conditions, yet they were more likely to have delayed or foregone needed healthcare. Midlife adults with MCC (with or without SPD) tended to report fewer barriers to healthcare and had higher past year health services use, suggesting that those with chronic conditions may be more connected to the healthcare system because of their chronic conditions. Yet, they still had significantly more barriers to accessing needed care compared to those with neither SPD nor MCC. In contrast, those with SPD (with or without MCC) had significantly higher past year emergency room visits and hospitalizations than those with neither. This might suggest an increase use of ERs and hospitalizations for those with mental distress as a result of their significantly higher odds of unmet healthcare needs.

Previous studies on mental health, multiple chronic conditions, and access to healthcare report mixed findings. We found significantly less access to mental healthcare by multimorbidity status, which is consistent with previous studies on access among those with SPD or serious mental illness (Alang, 2015; Mojtabai, 2009; Walker, Cummings, Hockenberry, & Druss, 2015; Weissman et al., 2017) and those with comorbid mental and physical health conditions (Jolles, Haynes-Maslow, Roberts, & Dusetzina, 2015). In contrast, other studies found that those with SPD were more likely to overuse healthcare (Fogarty, Sharma, Chetty, & Culpepper, 2008; Weissman et al., 2017). Specifically, those with SPD were more likely to have excess healthcare visits and emergency department visits due to inaccurate self-assessment of symptoms and their associated need for healthcare (Fogarty et al., 2008).

Our findings indicate that the majority of midlife adults had seen a general doctor in the past year, which is encouraging. However, adults with SPD/MCC were significantly more likely to report needing but not getting medical care, potentially creating a paradoxical situation where they use certain services such as general doctors more frequently than adults with neither condition, but are unable to access other services that would truly address their needs, such as prescription medication and follow-up care. It may also be a function of the inability to afford recurring out-of-pocket expenses and the need for ongoing medical care to manage chronic conditions or ongoing psychotherapy. Since health insurance is associated with employment for people under 65, the high unemployment rates and the loss of health insurance among those with SPD and MCC further exacerbates the potential unmet need for health care including mental health services.

Differences in unmet healthcare needs may be driven in part by social and economic characteristics of those with mental health and other chronic conditions. Similar to other studies, we found that those with SPD (with or without MCC) were characterized by considerable social disadvantage, such as low educational attainment, unemployment, and poverty, which are all associated with health status and access to care (Kim, Ford, Chiriboga, & Sorkin, 2012; Walker, Cummings, et al., 2015; Weissman et al., 2017). Accordingly, we found cost to be a significant barrier to receiving needed care for all SPD and MCC groups, compared to those with no SPD or MCC, even after adjusting for predisposing, enabling, and need factors. In our study, those with both SPD and MCC were the most adversely affected, followed by those with SPD alone, MCC alone, and neither. In contrast to the sociodemographic patterns where those with both SPD and MCC were the most disadvantaged, midlife adults with SPD alone were the most likely to be uninsured and to have no usual source of care. All midlife adults with MCC, even those with SPD, were more likely to have a usual source of care than those that did not have MCC, which may be due to ongoing chronic condition care. In spite of healthcare policy reform, public health program expansions, and other care models, cost remains a barrier to receiving needed healthcare for managing multimorbid mental and chronic conditions for a large proportion of midlife adults.

The most substantial differences in reasons for delaying care by multimorbidity status were structural barriers such as office hour availability, difficulty getting through on the phone, and inability to make a timely appointment. This suggests that for adults with SPD alone or SPD/MCC, navigating the healthcare delivery system is a challenge in and of itself. Those with severe psychological distress are seemingly those least able to advocate for themselves and to manage the burden of treatment in order to get what they need (May et al., 2014). Moreover, those with SPD may be overusing ERs and hospitalizations due to these structural barriers that result in their inability to access the outpatient system in a timely manner. Structural barriers may result in part from the larger proportion of adults with SPD/MCC who are publicly insured, compared to adults with neither. While wait times for appointments have decreased after the PPACA’s Medicaid expansion provisions were implemented, individuals insured under Medicaid nevertheless remain less able to make timely appointments compared to those who are privately insured, especially in states where Medicaid reimbursement rates remain low (Miller & Wherry, 2017; Polsky et al., 2015).

Differences in indicators of health status such as fair or poor health, chronic conditions, and functional limitations by multimorbidity status were striking. In our midlife sample, 69% of midlife adults had one or more chronic conditions, which is the same as the national US rate (Smolka et al., 2009). However, while one-third of all midlife adults had MCC, over 70% of midlife adults with SPD had MCC. The 20% prevalence of activity limitations we found is consistent with the 20% all midlife adults reported previously (Smolka et al., 2009). However, activity limitations were profoundly higher among those with any SPD and somewhat higher among those with MCC alone compared to those with neither SPD nor MCC. Importantly, activity limitations are associated with higher healthcare costs, especially when combined with chronic conditions (Hayes, Salzberg, & McCarthy, 2016). This further underscores the need for targeted healthcare to monitor lifestyle and risk factors, provide early detection and management of chronic conditions, and address mental health needs before comorbid physical and mental health conditions lead to functional or activity limitations and reduced quality of life.

Implications for policy and/or practice

Reducing barriers to access for all types of healthcare, including mental health services, for midlife adults with mental distress and chronic conditions must be prioritized by both policymakers and healthcare providers. The most common reported barrier to care was cost, perhaps due to uninsurance or underinsurance, high deductible health plans, or lack of in-network providers. This is particularly concerning given limited networks that accept Medicaid (Decker, 2012) or Medicare assignment (U.S. Centers for Medicare & Medicaid Services, 2017), which are forms of public insurance the majority of our midlife population with SPD/MCC had. Lowering the age of eligibility for Medicare into the midlife range could provide coverage that some midlife adults do not have (Moon & Uccello, 2020). However, more than one-third of mental health providers are not covered by Medicare and current Medicare enrollees describe considerable barriers to mental health treatment caused by the Medicare mental health coverage gap (MMHCG) (Fullen, Dolbin-MacNab, Brossoie, Wiley, & Lawson, 2020; Fullen, Wiley, & Morgan, 2019). Cost-sharing for mental health services under Medicare (20%) may also be a barrier to individuals with severe mental illness (Chaiyachati, Livesey, & Liao, 2020), which is likely increased for those with both SPD and MCC (Chaiyachati et al., 2020). Efforts to address both the MMHCG and the cost-sharing for mental health services under Medicare will also be needed for Medicare expansion to be a significant policy solution.

Efforts to expand Medicaid access and enrollment may be also effective in increasing use of both physical and mental health services among midlife adults prior to reaching Medicare-eligible age. Redirecting the focus of Section 1115 Medicaid demonstration waivers away from work requirements may be particularly effective in serving adults with MCC/SPD, alone or in combination, as either condition may present a barrier to work (Kaiser Family Foundation, 2021; Sommers, Goldman, Blendon, Orav, & Epstein, 2019; Westmoreland, Bloche, & Gostin, 2021). Continued expansion of Medicaid eligibility in additional states to adults without dependent children may also ensure access to care, especially in low-income vulnerable adults (Kino & Kawachi, 2018; Lee & Porell, 2020).

The second most common cited barrier was the inability to secure a timely appointment. Although, this likely refers to all types of care, it may be especially true for mental health services due to the documented shortage of mental health providers (Mechanic & Olfson, 2016; Olfson, 2016). Incentives to increase the overall number of mental health providers could reduce extended waits for appointments. Increased accessibility of mental health providers could be achieved by expanding the range and types of providers that are covered in-network and diversifying mental health service settings (e.g. primary care clinics, community settings, and telehealth). Telehealth, in particular, has had a dramatic increase in use for mental health visits during the COVID-19 pandemic, demonstrating the feasibility of this mode of delivery beyond rural settings (Pfender, 2020). Yet, payers are not required to reimburse telehealth visits at the same rate as in-person care (Warren & Smalley, 2020). Parity laws would ensure comparable payment for telehealth as for in-person services. However, telehealth alone does not address the ongoing shortage of mental health providers.

Ensuring access to early and adequate screening and treatment for both physical and mental health conditions during midlife is essential to ease the burden of disease and unhealthy aging. Additionally, it will relieve an overloaded healthcare system and reduce excess healthcare costs (Mechanic & Olfson, 2016). Primary care providers may lack necessary training or available time to address serious mental illness. Likewise, psychiatrists lack training and time to manage or monitor chronic conditions. Mental health treatments have recently been shifting toward pharmacological interventions, which has implications for management of comorbid psychiatric and chronic conditions. This is an especially vulnerable population as many psychotropic medications require close monitoring of cardiometabolic risk factors (Abosi, Lopes, Schmitz, & Fiedorowicz, 2018; American Diabetes Association, American Psychiatric Association, American Association of Clinical Endocrinologists, and North American Association for the Study of Obesity, 2004). However, in spite of formal guidelines, monitoring of metabolic risk by prescribing psychiatrists remains inadequate (Nasrallah, 2012). Similarly, it is recommended that primary care providers routinely screen patients for depression, yet the overall screening rate remains low (Akincigil & Matthews, 2017). Co-located care has received some attention as one potential solution to co-managing the disconnect between physical and mental health care (Kilbourne et al., 2011; Pirraglia, Kilbourne, et al., 2011).

Limitations

Our findings should be considered in light of potential limitations. First, the K6 measures non-specific psychological distress by assessing symptoms of depression and anxiety, and thus may not capture the full range of mental illnesses that can prevent adequate access to healthcare in midlife adults. However, the K6 measures self-report of symptoms, thus it may be more likely to capture those with current or recent psychological distress than self-reports of a clinical diagnosis. Additionally, we found that 13% of this population had moderate mental distress, which were included in the no distress category. This likely underestimates the magnitude of unmet care for those with SPD. Second, our measure of MCC is based on self-reports of chronic conditions. While self-reports are subject to misclassification, these questions ask whether a healthcare professional has ever told them they had each condition, which may mitigate over-reporting. Third, NHIS questions about foregone care ask about needing but not getting care due to cost. Midlife adults with MCC and/or SPD may have other reasons for unmet healthcare needs. Unmet needs may have been underestimated due to the emphasis on cost as the barrier. Finally, the recall periods for chronic conditions are ever in one’s life, for symptoms of psychological distress is 30 days, and for delayed and foregone care is the past year. Consequently, it is possible the time periods of delayed or foregone care and psychological distress do not overlap for some individuals. However, chronic conditions by definition are ongoing, and serious mental illness, while often episodic, can be disabling even when in remission.

Conclusions

Midlife adults with SPD and MCC are severely debilitated and have substantial unmet healthcare needs. Midlife is a critical window of opportunity for identifying and managing both chronic conditions and mental illness, in order to facilitate healthy aging and increase quality of life. Despite recent policy advances such as the Affordable Care Act, cost remains a major reason for delaying or foregoing care, especially among midlife adults with SPD and/or MCC. Lowering the age for Medicare eligibility, continuing Medicaid expansion to additional states, and decreasing out of pocket expenses, may help address cost barriers. The ongoing inability to get timely appointments could be mitigated by addressing the mental health services workforce shortage and increasing the accessibility of mental health providers through expanding the types of providers that are covered in-network and diversifying mental health service settings, including reimbursement for telehealth visits. Approaches to both health policy and clinical practice must recognize that midlife adults with psychological distress may have difficulty navigating the healthcare system or advocating for themselves, further preventing them from accessing needed care. Additional coordinated efforts by policymakers and healthcare providers are crucial to address the complex healthcare needs of this population at a critical stage of the life-course.