Association of mental health outcomes and lower patient satisfaction among adults with alopecia: A cross-sectional population-based study

Abstract

Background

Previous studies have found the increasing use of patient satisfaction scores by patients and insurance payers. Less is known about how patient mental health affects health care satisfaction.

Objective

To examine the association between baseline mental health and health care satisfaction among adults with alopecia.

Methods

We examined 543 adults with alopecia in the 2004-2016 Medical Expenditure Panel Survey. Mental health burden was assessed by the 6-item Kessler Psychological Distress Scale (K6) and 2-item Patient Health Questionnaire (PHQ2). Patient satisfaction was determined using the Consumer Assessment of Healthcare Providers and Systems survey.

Results

Adults with versus without alopecia had higher rates of positive PHQ2 (adjusted odds ratio [95% CI], 1.37 [1.05-1.78]); positive K6 (1.57 [1.02-2.41]), and comorbid anxiety (1.85 [1.30-2.63]) and depression (1.68 [1.19-2.39]). Positive PHQ2 (2.15 [1.13, 4.11]) and positive K6 (6.04 [2.60, 14.05]) were associated with low patient satisfaction. Whereas, there were no differences in the rates of low patient satisfaction associated with comorbid anxiety (0.74 [0.33-1.67]) and depression (1.42 [0.72-2.78]).

Limitations

Data are unavailable on alopecia areata phenotypes and treatment.

Conclusions

Adults with alopecia and greater mental health symptoms report lower patient satisfaction. Clinicians may wish to adapt their communication style to support these patients and improve overall health care satisfaction.

Capsule Summary.

• •Patients with alopecia and comorbid depressive symptoms and/or psychological distress reported lower patient satisfaction, compared with alopecia patients without mental health symptoms.
• •Patients with alopecia and mental health symptoms may benefit from individualized communication strategies aimed at supporting the patient.

Introduction

Patient satisfaction scores are increasingly used by patients, administrators, and insurance to evaluate clinician performance.^1,2 Dermatologists generally have high patient satisfaction scores, and previous studies found that physician characteristics such as years of experience are associated with higher patient satisfaction with dermatologists.^3,4 However, it is not well-understood why some patients give low satisfaction scores to dermatology clinicians whom many other patients rate highly. Overall, data are limited regarding the association of patient characteristics, such as baseline psychological status, with patient satisfaction. Patients with alopecia areata are at higher risk of depression and anxiety, making these patients a particularly important group for research.^5,6 These findings are relevant for clinical care because higher patient satisfaction is associated with improved patient outcomes.⁷ Understanding the potential association between mental health outcomes and patient satisfaction empowers clinicians to adapt patient interactions and address psychological health to improve health care satisfaction. As such, the objective of the present study was to characterize mental health outcomes among adults with alopecia and examine the association of patient satisfaction with mental health symptoms and diagnoses.

Methods

This was a cross-sectional analysis of adults (≥18 years) enrolled in the 2004-2016 Medical Expenditure Panel Survey (MEPS), annual surveys of health functioning and status conducted by the Agency for Health Research and Quality (AHRQ) of the US noninstitutionalized population. The MEPS includes data on sociodemographic characteristics, health status, and patient satisfaction. The AHRQ provided complex survey weights, strata, and clustering to generate representative nationwide estimates of health status/function. Survey weights were constructed to adjust for nonresponse and adjusted by iterative proportional fitting to population estimates for the corresponding year using age, sex, race and ethnicity, geographic region, and metropolitan area status.

The study cohort included all adults (≥18 years) with a current diagnosis of alopecia areata and related hair conditions. Respondents were asked to list all their current clinical diagnoses. Diagnoses were transcribed verbatim and converted into International Classification of Diseases, Clinical Modification codes by professional coders. Alopecia diagnosis was identified by International Classification of Diseases, Clinical Modification codes and compared with controls without a diagnosis of alopecia.

Patient satisfaction

The primary study outcome was patient satisfaction. MEPS evaluated satisfaction using the validated Consumer Assessment of Healthcare Providers and Systems (CAHPS) survey (range, 4-16).⁸ The specific questions are reproduced in Supplementary Table I (available via Mendeley at https://doi.org/10.17632/59xns3ckdy.1). Patient satisfaction was stratified into low satisfaction (CAHPS score ≤9), moderate satisfaction (CAHPS score 10-15), and high satisfaction (CAHPS score 16), consistent with stratification by previous analyses.⁹

Psychological state and covariates

Psychological status was measured using the 6-item Kessler Psychological Distress Scale (K6) and 2-item Patient Health Questionnaire (PHQ2). The K6 contains 6 items that ask about feelings of nervousness, hopelessness, restlessness, depression, apathy, and worthlessness over the last 30 days (range, 0-24). K6 scores were stratified into no to mild (0-12) and clinically significant psychological distress (≥13). This threshold was selected because a K6 score ≥13 has been associated with clinically significant psychological distress.¹⁰ PHQ2 screens for depressive symptoms by the frequency of depressed mood and anhedonia over the most recent 2 week period (range, 0-6). Previous research found that PHQ2 ≥2 diagnosed major depression with 86% sensitivity and 78% specificity, and this cut-off was considered a positive screen for depression in this study.¹¹

Statistical analysis

Prevalence and 95% CI of alopecia were calculated, stratified by sex, age, income, race and ethnicity, insurance payer, and education. Pearson χ2 tests with Rao-Scott adjustment were used to assess differences. Logistic regression models were constructed to evaluate the association of alopecia diagnosis with PHQ2, K6, depression, and anxiety. Additionally, Cochrane-Armitage tests were used to examine the change in the rates of mental health outcomes among adults with alopecia over the study period (2004-2007/2008-2012/2013-2016).

Median and IQR of CAHPS score were calculated, stratifying by mental health measure: PHQ2, K6, depression, and anxiety. Linear regression models were constructed to examine the association between mental health outcomes (independent variable) and CAHPS score (dependent variable) among adults with alopecia. Least-squares means were calculated; adjusted β with 95% CI were reported. We also used multivariable logistic regression models to determine the impact of positive K6 and PHQ2 screen and comorbid depression and anxiety on rates of low patient satisfaction (CAHPS score ≤9). Adjusted odds ratios (OR) and 95% CI were reported.

All regression models were adjusted for sociodemographics and functional and health status, as recommended by the AHRQ to isolate the effect of patient-provider interaction.¹² Covariates included sociodemographic characteristics (sex [male/female], age [continuous], race and ethnicity [White/Black/Hispanic/other or multiple], education attainment [less than high school//high school degree], income as a percentage of federal poverty level, and insurance payer [private/public/none]), and multimorbidity. Multimorbidity burden was assessed by the Charlson Comorbidity Index, a comorbidity scoring system calibrated to estimate 1-year mortality.^13,14

Statistical analyses were conducted in SAS v9.4 (SAS Institute). A 2-sided P value <.05 was considered statistically significant. The Northwestern University Institutional Review Board determined that this study did not constitute human subjects research because all data were deidentified and publicly available from the AHRQ. This study complies with all Strengthening the Reporting of Observational Studies in Epidemiology recommendations.¹⁵

Results

Population characteristics

Overall, there were 178,161 US adults ≥18 years who were surveyed by the MEPS and completed the CAHPS survey, including 543 (prevalence, 0.3%) adults with alopecia. Alopecia was associated with younger age, higher household income, private insurance coverage, and having a high school degree (Rao-Scott χ², P ≤ .01 for all) (Table I).

Table I.

Sociodemographic characteristics of us adults with and without alopecia (n = 178,161)^∗

Characteristic	Alopecia Diagnosis
	N = 177,618		Yes (n = 543)
	Raw frequency	Weighted prevalence [95% CI]	Raw frequency	Weighted prevalence [95% CI]	Rao-Scott P value
Sex					.42
Male	70,390	42.5 [42.2-42.8]	198	40.0 [34.1-45.9]
Female	107,228	57.5 [57.2-57.8]	345	60.0 [54.1-65.9]
Age (y)					.01
18-39	58,497	32.2 [31.6-32.9]	208	39.4 [32.6-46.2]
40-59	66,062	36.7 [36.2-37.2]	217	37.1 [31.0-43.1]
≥60	53,059	31.1 [30.3-31.8]	118	23.5 [19.1-28.0]
Income					.0001
Poor, near poor, low	64,786	25.9 [25.2-26.6]	127	16.9 [13.1-20.7]
Middle income	51,441	29.1 [28.6-29.6]	152	26.9 [21.7-32.1]
High income	61,391	45.0 [44.1-45.9]	264	56.1 [49.7-62.6]
Race/ethnicity					.19
White	95,850	72.4 [71.2-73.6]	325	76.2 [71.9-80.4]
Black	31,806	10.5 [9.7-11.3]	98	10.5 [7.7-13.3]
Multiracial/other	14,627	6.6 [5.9-7.3]	43	5.2 [3.1-7.4]
Hispanic	35,335	10.5 [9.7-11.4]	77	8.1 [5.7-10.5]
Insurance coverage					<.0001
Private	114,488	73.6 [72.9-74.4]	430	85.0 [81.2-88.9]
Public	45,142	19.0 [18.4-19.7]	92	12.7 [9.0-16.5]
None	17,988	7.3 [7.0-7.6]	21	2.2 [0.9-3.5]
Education					<.0001
Less than HS	35,607	13.6 [13.1-14.0]	57	6.3 [4.3-8.3]
HS degree	140,815	86.4 [86.0-86.9]	486	93.7 [91.7-95.7]

HS, High school.

^∗Bold values indicate statistically significant P values. Rao-Scott χ² tests were used to assess sociodemographic differences among adults with versus without alopecia.

Alopecia and mental health outcomes

Compared to those without alopecia, adults with alopecia had higher rates of positive PHQ2 screening (adjusted OR [95% CI], 1.37 [1.05-1.78]; P = .02), positive K6 screening (1.57 [1.02-2.41]; P = .04), and comorbid anxiety (1.85 [1.30-2.63]; P = .0007) and depression (1.68 [1.19-2.39]; P = .004) in logistic regression models adjusted for age, sex, race and ethnicity, education, income, insurance coverage, and multimorbidity.

Patient Health Questionnaire-2 depressive symptoms

Overall, 75.5% of adults with alopecia had a negative PHQ2 screen, and 24.5% had a positive PHQ2 screen for depressive symptoms. The median [IQR] CAHPS was 14.5 [12.0, 15.4] and 13.2 [11.2, 15.2] for those with negative and positive PHQ2 screening, respectively.

Compared to those with a negative PHQ2 screening, adults with alopecia and positive PHQ2 had lower satisfaction (LS-means, 13.52 vs 14.29; adjusted β [95% CI], −0.77 [−1.17, −0.37]; P = .0002). Moreover, positive PHQ2 screening was associated with higher rates of low patient satisfaction (CAHPS ≤9) (proportion, 10.7% vs 4.9%; aOR [95% CI], 2.15 [1.13, 4.11]; P = .02) (Table II).

Table II.

Mental health, sociodemographic, and clinical associations of low patient satisfaction among adults with alopecia (n = 543)^∗

Characteristic	Psychological distress		Depressive symptoms		Anxiety		Depression
	Adjusted OR [95% CI]	P value	Adjusted OR [95% CI]	P value	Adjusted OR [95% CI]	P value	Adjusted OR [95% CI]	P value
Mental health outcome
Negative screen/no	1.00 [ref]	-	1.00 [ref]	-	1.00 [ref]	-	1.00 [ref]	-
Positive screen/yes	6.04 [2.60-14.05]	<.0001	2.15 [1.13-4.11]	.02	0.74 [0.33-1.67]	.46	1.42 [0.72-2.78]	.31
Sex
Male	1.00 [ref]	-	1.00 [ref]	-	1.00 [ref]	-	1.00 [ref]	-
Female	2.80 [1.64-4.80]	.0003	2.98 [1.87-4.73]	<.0001	2.59 [1.59-4.24]	.0002	2.52 [1.52-4.17]	.0005
Age (y)	0.97 [0.95-0.99]	.003	0.97 [0.96-0.99]	.007	0.98 [0.96-0.99]	.004	0.97 [0.96-0.99]	.003
Education
No HS diploma	1.00 [ref]	-	1.00 [ref]	-	1.00 [ref]	-	1.00 [ref]	-
HS diploma	1.87 [0.76-4.59]	.17	2.40 [0.98-5.87]	.05	2.43 [0.97-6.07]	.06	2.23 [0.92-5.41]	.08
Income
Poor, near poor, low income	0.69 [0.36-1.35]	.28	0.86 [0.47-1.57]	.61	0.92 [0.51-1.66]	.78	0.85 [0.47-1.53]	.59
Middle income	0.79 [0.38-1.61]	.50	0.74 [0.34-1.61]	.44	0.83 [0.42-1.67]	.60	0.82 [0.40-1.68]	.58
High income	1.00 [ref]	-	1.00 [ref]	-	1.00 [ref]	-	1.00 [ref]	-
Race/ethnicity
White	1.00 [ref]	-	1.00 [ref]	-	1.00 [ref]	-	1.00 [ref]	-
Black	0.27 [0.12-0.61]	.002	0.30 [0.14-0.67]	.004	0.28 [0.12-0.63]	.003	0.28 [0.13-0.63]	.003
Multiracial/other	0.40 [0.23-0.72]	.003	0.51 [0.29-0.90]	.02	0.50 [0.27-0.94]	.03	0.53 [0.28-1.01]	.05
Hispanic	0.43 [0.17-1.09]	.07	0.49 [0.18-1.37]	.17	0.45 [0.15-1.36]	.15	0.46 [0.16-1.38]	.16
Insurance coverage
Private	1.00 [ref]	-	1.00 [ref]	-	1.00 [ref]	-	1.00 [ref]	-
Public	1.68 [0.85-3.32]	.13	1.85 [0.95-3.59]	.07	2.14 [1.07-4.26]	.03	1.95 [0.95-3.99]	.07
None	1.42 [0.33-6.11]	.64	1.83 [0.46-7.27]	.39	2.04 [0.46-9.01]	.34	2.05 [0.51-8.27]	.31
Charlson Comorbidity Index	0.69 [0.44-1.09]	.11	0.81 [0.54-1.23]	.32	0.82 [0.56-1.21]	.31	0.81 [0.55-1.20]	.29

HS, High school.

^∗Bold values indicate statistically significant P values. Multivariable logistic regression models were constructed to examine associations with low patient satisfaction (dependent variable) among adults with alopecia. Model covariates included sociodemographic characteristics and Charlson Comorbidity Index scores.

Kessler-6 psychological distress

Among adults with alopecia, 93.5% had no to mild and 6.5% had clinically significant psychological distress. The median [IQR] CAHPS was 14.3 [11.8, 15.4] for those with no to mild distress and 12.6 [10.7, 14.9] for those with clinically significant distress.

Compared with no to mild psychological distress, clinically significant psychological distress was associated with decreased patient satisfaction (LS-means, 12.85 vs 14.23; adjusted β [95% CI], −1.38 [−2.14, −0.63]; P = .0005) among adults with alopecia after adjustment for sociodemographics and multimorbidity. Further, clinically significant psychological distress was associated with 6-fold higher odds of low patient satisfaction (18.7% vs 5.8%; aOR [95% CI], 6.04 [2.60, 14.05]; P < .0001) (Table II).

Comorbid depression and anxiety

Of 543 adults with alopecia, 105 (17.9%) had comorbid depression and 84 (17.3%) had anxiety. There was no association between comorbid depression and patient satisfaction scores (LS-means, 13.85 vs 14.06; adjusted β [95% CI], −0.21 [−0.69, 0.28]; P = .40). Anxiety was associated with lower patient satisfaction (13.54 vs 14.06; −0.53 [−1.02, −0.03]; P = .04).

There were no differences in the rates of low patient satisfaction associated with comorbid anxiety (0.74 [0.33-1.67]; P = .46) and depression (1.42 [0.72-2.78]; P = .31).

Discussion

Overall, this study found that alopecia is associated with increased mental health symptoms, and psychological distress and depressive symptoms were associated with lower overall patient satisfaction among adults with alopecia. These findings confirm and build on previous studies that found increased mental health burden associated with alopecia areata and highlight the particular relevance of mental health and patient satisfaction for these patients.^5,6 Specifically, adults with positive PHQ2 screening for depressive symptoms had 2.2-fold increased odds and positive K6 had 6-fold higher odds of low patient satisfaction, whereas neither comorbid depression nor anxiety was associated with low patient satisfaction. These findings suggest that the underdiagnosis of mental health symptoms among adults with alopecia may contribute to low patient satisfaction, and/or poor satisfaction with health care may predispose to mental health symptoms. Clinicians may wish to consider screening alopecia patients for mental health symptoms to identify those at particular risk of low patient satisfaction.

We found that increased mental health burden was associated with lower patient satisfaction among adults with alopecia, regardless of sociodemographic characteristics and comorbidities. These results are consistent with previous studies that found that comorbid depression and anxiety are associated with lower patient satisfaction in elderly adults.16, 17, 18 There are multiple explanations for this association. Psychological state may influence how patients perceive clinician communication, regardless of the quality of clinical care. Increased baseline psychological distress and depression may predispose to suboptimal interactions with providers. Additionally, low health care satisfaction in the setting of a chronic condition may worsen mental health symptoms. There may be a cyclical relationship such that comorbid mental health symptoms worsen patient satisfaction, and low patient satisfaction contributes to the patient's mental health burden. The cross-sectional design of this study precluded any examination of causation; longitudinal studies are needed to assess the directionality of the association.

Interestingly, positive screening on PHQ2 for depressive symptoms or K6 for psychological distress were associated with lower patient satisfaction, yet comorbid depression and anxiety were not associated with differences in patient satisfaction. Together, it may be that subclinical, undiagnosed mental health symptoms predispose to low patient satisfaction. Moreover, dissatisfaction with health care experiences may also lead to mental health symptoms. The association between alopecia and increased mental health burden suggests a need for clinicians to be vigilant in screening for and managing mental health symptoms. Previous studies have shown the benefits of mindfulness and cognitive behavior therapy among patients with alopecia.^19,20 As such, this study builds on previous research that multidisciplinary, team-based care to ensure comprehensive management of patient needs may also improve patient satisfaction among adults with alopecia.^21,22

Our findings have important clinical implications for care of patients with alopecia. Improving the patient experience may improve clinical outcomes because highly satisfied patients are more likely to continue with the same provider, share information with their physician, and adhere to treatment plans.⁷ Moreover, clinicians should recognize the effect of mental health on satisfaction and adapt their communication accordingly. Depressed or anxious patients with alopecia may benefit from a more supportive communication style and shared decision-making in treatment decisions.²³

Since 2012, the Center for Medicare and Medicaid Services introduced CAHPS questions as a metric in the calculation of reimbursement.^2,24 Given the increased mental health burden of alopecia, our study provides data suggesting that the use of CAHPS in the calculation of reimbursement for alopecia is problematic. In particular, comparing clinician performance based on patient satisfaction scores may asymmetrically penalize clinicians who care for alopecia patients with comorbid mental health symptoms. Additional research is needed to adjust satisfaction scores for the patient psychological baseline state.

The strengths of this study include the complex survey design that allowed for representative estimates of the US population and data collection over a 13-year period. However, some limitations merit mention. First, data were unavailable on patterns of hair loss and treatment regimens. Also, our data were limited to US adults, and the results cannot be generalized to international patients. A cross-sectional study of elderly adults in the Netherlands found that late-life depression was associated with lower health care satisfaction.¹⁶ Similarly, psychiatric symptom improvement was associated with higher levels of satisfaction among involuntary inpatients across 22 hospitals in England.²⁵ However, neither study specifically examined satisfaction among patients with alopecia areata. Future studies in other countries are warranted to validate these findings among patients with alopecia on an international scale. These results are mostly generalizable to dermatologists because most patients with alopecia areata have their disease managed by a dermatologist. Last, the cross-sectional study design precluded any conclusions about causality or directionality.

In conclusion, adults with alopecia and psychological distress or depressive symptoms are more likely to report low patient satisfaction compared with adults with alopecia without mental health symptoms. Lower patient satisfaction among those with positive mental health screening and not comorbid psychiatric disorders suggests that the underdiagnosis and undermanagement of mental health symptoms may contribute to lower patient satisfaction. Clinicians should recognize that baseline mental health symptoms may affect patient satisfaction, and patients may benefit from tailored communication. Further research is necessary to determine the optimal strategies to improve satisfaction in adults with alopecia and mental health symptoms.

Conflicts of interest

None disclosed.