Associations between demographic characteristics and unmet supportive care needs in adults with cystic fibrosis

Abstract

Context

Patients living with cystic fibrosis (CF) report impaired quality of life. Little is known about unmet supportive care needs among adults living with CF and how they are associated with demographic characteristics.

Objectives

The primary objective of this study was to identify associations between demographic variables and unmet supportive care needs regarding anxiety, sadness, pain and uncertainty about the future of living with CF.

Methods

We recruited 165 adults with CF from a single academic medical centre to complete a brief demographic survey and the Supportive Care Needs Survey (SCNS-34), a validated self-reported needs assessment that measures the prevalence of and preferences for support for 34 needs that commonly occur in patients with serious illness.

Results

Approximately half of the participant sample was male, with a median age of 29 years, varying income levels and a range of lung disease severity. We found statistically significant associations between insufficient income and increased odds of reporting need for support regarding anxiety (OR: 6.48; 95% CI 2.08 to 20.2), sadness (OR: 6.15; 95% CI 2.04 to 18.5), pain (OR: 7.06; 95% CI 2.22 to 22.4) and worries surrounding uncertainty about the future (OR: 3.43; 95% CI 1.18 to 9.99).

Conclusion

Adults with CF report significant unmet needs for support in several physical and emotional domains. Many of these domains were associated with demographic characteristics, most notably, income. Our findings underscore the importance of developing treatment approaches that are sensitive to patient demographics when addressing unmet supportive care needs among adults with CF.

INTRODUCTION

Cystic fibrosis (CF) affects more than 72 000 individuals and their families worldwide, and patients with CF often report physical and emotional burdens that impair their quality of life.¹ Yet, little is known how patient characteristics, such as demographics, are associated with unmet supportive care needs in CF. Among a sample of adults with CF, we investigated associations between demographic variables and preferences for support with four specific needs that are common among individuals with life-limiting illness²: uncertainty about the future, anxiety, sadness and pain management.

METHODS

We recruited a convenience sample of adults with CF from the University of Pittsburgh Medical Center (UPMC) adult CF programme to complete the Supportive Care Needs Survey (SCNS-34). Eligibility criteria were age ≥ 18 years and the ability to respond in English. Data were collected as part of a quality improvement project, approved by the UPMC Quality Improvement Review Committee.

Data were collected at point of care using a tablet computer. We collected clinical information from the electronic medical record, including forced expiratory volume in 1 s, body mass index, lung transplant status, CF-related diabetes, pulmonary exacerbations and hospitalisations in the prior 12 months. Participants self-reported health status as excellent, good, fair or poor, as well as demographics (eg, education, religiosity, relationship status and income).

The SCNS-34 is a patient-reported assessment of the presence of and need for support with each of 34 items commonly experienced by individuals with serious illness.³ Originally developed in cancer, the SCNS-34 has been successfully modified and used in other serious illnesses, including heart failure.⁴ Key dependent variables of interest were: need for support with anxiety, sadness, pain and uncertainty about the future. We chose these four needs due to their high prevalence in other seriously ill populations, and as they are often the target of palliative interventions.⁵

Proportions and medians with ranges were calculated using Stata version 13. Multivariable logistic regression was used to identify associations between demographic variables and unmet supportive care needs. Separate models were fit for each unmet need. Our outcomes (unmet supportive care needs) were dichotomised for our primary analysis into two categories: ‘Any Need’ (including scores of ‘Low Need’, ‘Moderate Need’ and ‘High Need’ in the SCNS) and ‘No Need’. Several sensitivity analyses performed did not appreciably alter our results.

RESULTS

We surveyed 165 individuals, representing 71% of the UPMC adult CF programme population at the time of data capture. None of the participants were receiving specialist palliative care at the time of survey completion. Slightly more than half of the sample was male (n = 92), with a median age of 29 years (table 1). Fifty-eight per cent (n = 95) of participants noted religion as either fairly or very important. Twenty-eight per cent (n = 46) indicated some need for support regarding sadness, 39% (n = 64) for anxiety, 39% (n = 64) for uncertainty about the future and 37% (n = 61) for pain. Half of our participants (n = 82) rated their health as poor or fair.

Table 1.

Demographic and clinical characteristics of a convenience sample of adults with cystic fibrosis (N = 165)

Characteristic	n (%)
Pulmonary impairment
Mild (FEV1 ≥80%)	42 (25)
Moderate (50%–79%)	58 (35)
Severe or very severe (<49%)	65 (39)
Gender
Male	92 (56)
Female	73 (44)
Race
Caucasian	161 (98)
African-American	1 (1)
Other or multiple races	3 (2)
Religiosity (importance of religion)
Not at all important	34 (21)
Not too important	36 (22)
Fairly important	46 (28)
Very important	49 (30)
Religious affiliation
Protestant Christian	30 (18)
Roman Catholic	44 (27)
Other Christian	30 (18)
Jewish	2 (1)
Buddhist	1 (1)
Agnostic/atheist/no religion	21 (13)
Other	37 (22)
Age in years
Median (range)	29 (18–66)
18–29	84 (51)
30–39	56 (34)
40–49	14 (8)
50+	11 (7)
Education
Some high school or less	4 (2)
High school diploma/GED	52 (32)
Vocational school	2 (1)
Some college	32 (19)
College degree	58 (35)
Professional or graduate degree	17 (10)
Financial status
Money left over	75 (45)
Just enough money to make ends meet	69 (42)
Not enough money to make ends meet	21 (13)
Relationship status
Single	102 (62)
Partnered/married	63 (38)
BMI (kg/m2)
Median (range)	21.3 (15.2–43.0)
Transplant status
Not evaluated or listed	159 (96)
Awaiting transplant	5 (3)
Post-transplant	1 (1)
CF-related diabetes
No	100 (61)
Yes	65 (39)
Pulmonary exacerbations in prior 12 months
Median (range)	0 (0–21)
Hospitalisations in prior 12 months
Median (range)	1 (0–36)
Health status
Excellent	20 (12)
Good	62 (38)
Fair	63 (38)
Poor	20 (12)

BMI, body mass index; CF, cystic fibrosis; FEV₁, forced expiratory volume in 1 s.

From the 39% of respondents who reported some need for support with anxiety, we found that participants who reported insufficient income and were older than 29 years presented higher odds of reporting need. In our sample, males had lower odds of reporting need for support with anxiety (table 2).

Table 2.

Adjusted ORs from multivariable logistic regression on prevalence of unmet supportive care needs among a convenience sample of adults with cystic fibrosis^*

Variable		Anxiety aOR (95% CI)	Sadness aOR (95% CI)	Pain aOR (95% CI)	Uncertainty about the future aOR (95% CI)
Pulmonary impairment	Mild (FEV1 ≥80%) (ref.)
	Moderate (50%–79%)	0.80 (0.31 to 2.07)	0.61 (0.22 to 1.68)	2.71 (0.97 to 7.58)	1.19 (0.47 to 3.01)
	Severe or very severe (<49%)	1.49 (0.55 to 4.02)	1.57 (0.56 to 4.43)	2.90 (0.96 to 8.79)	2.60 (0.96 to 7.06)
Age	18–29 (ref.)
	30–39	2.37 (1.02 to 5.51)*	1.43 (0.61 to 3.39)	1.61 (0.67 to 3.88)	1.08 (0.48 to 2.42)
	40–49	4.35 (1.05 to 18.0)*	2.61 (0.57 to 11.9)	10.25 (2.16 to 48.6)*	0.75 (0.17 to 3.26)
	50+	4.51 (1.04 to 19.7)*	2.97 (0.64 to 13.7)	4.57 (0.97 to 21.5)	2.42 (0.58 to 10.1)
Male gender		0.42 (0.20 to 0.88)*	0.63 (0.29 to 1.38)	0.62 (0.29 to 1.32)	0.62 (0.31 to 1.27)
Religiosity		0.70 (0.32 to 1.52)	0.41 (0.18 to 0.94)*	0.34 (0.15 to 0.78)*	0.63 (0.30 to 1.33)
Relationship status	Partnered	0.54 (0.23 to 1.28)	0.57 (0.23 to 1.41)	0.62 (0.26 to 1.52)	0.89 (0.40 to 2.01)
Income by the end of the month	Money left over (ref.)
	Just enough	2.19 (1.00 to 4.79)	1.30 (0.57 to 3.00)	2.77 (1.20 to 6.40)*	1.31 (0.62 to 2.77)
	Not enough	6.48 (2.08 to 20.2)*	6.15 (2.04 to 18.5)*	7.06 (2.22 to 22.4)*	3.43 (1.18 to 9.99)*
BMI (kg/m2)	(continuous)	1.02 (0.92 to 1.13)	1.03 (0.93 to 1.15)	1.00 (0.91 to 1.11)	1.06 (0.96 to 1.17)
Pulmonary exacerbations in prior year	(continuous)	1.11 (0.92 to 1.35)	1.00 (0.85 to 1.19)	1.24 (0.99 to 1.56)	1.21 (0.97 to 1.51)

Hosmer & Lemeshow’s goodness-of-fit test was used to confirm good model data fit (anxiety: 0.6335; sadness: 0.2117; pain: 0.3572; uncertainty: 0.8944).

^*Any needs=3, 4 or 5. An asterisk (*) in the data table indicates p value <0.05.

BMI, body mass index; FEV₁, forced expiratory volume in 1 s; aOR, adjusted OR.

Twenty-eight per cent of respondents reported some need for support regarding sadness. Among this subset, increased religiosity was associated with lower odds of reporting need for management of feelings of sadness, while participants who reported insufficient income by the end of the month had higher odds of reporting need for support with sadness (table 2).

Among the respondents who indicated need for support with pain management, adequate and insufficient income by the end of the month were both significantly associated with higher odds of reporting need. We also found that participants who reported increased religiosity were less likely to report need for support with pain management, and respondents between 40 years and 49 years were more likely to report overall need (table 2).

Thirty-nine per cent of respondents expressed need for support regarding uncertainty about the future. Only participants with insufficient income by the end of the month presented higher odds of reporting need for support for managing uncertainty (table 2).

CONCLUSION

Among a sample of adults with CF in the USA, we found associations between unmet psychological and physical supportive care needs and a variety of demographic characteristics.

Despite 28% and 39% of our sample reported having supportive care needs, these needs were not associated with pulmonary impairment. Our sample presented with a low average of pulmonary exacerbations in the prior year; as such, our findings are potentially an understatement of the association between pulmonary impairment and needs for supportive care. Furthermore, our work reinforces the prevailing notion that supportive services, namely palliative care, should not be restricted to individuals who are exclusively at the end of life, and screening of needs should be done systematically, proactively and throughout the disease course. Our results also underscore the importance of building capacity in primary palliative care, such that CF clinicians are better equipped to assess unmet supportive needs, address basic sources of suffering and refer to specialty services, such as palliative care, behavioural health and social work, when indicated.

Income was consistently associated with increased odds of reporting need for all four supportive care needs measured. As clinical models of palliative care in CF are developed, attention should be paid to factors that may serve as barriers, such as additional copayments (as is the case in the USA) given our findings. The sample in the present study was not seen by palliative care, yet these findings underscore the need for palliative care accessibility. Accessibility and affordability go hand in hand, especially in payer systems in which specialty care access involves additional cost.

The present study was performed in a large, urban academic CF centre that serves as a regional care centre, which may limit generalizability. Furthermore, the cross-sectional nature of our study limited our ability to determine causality and the observed wide CIs for some estimates may be due to small cell sizes for some variables. Lastly, our participants had higher education and income relative to patient populations at other US CF care centres; therefore, our findings may underestimate the true impact of insufficient income on supportive care needs.

Our findings demonstrate associations between a variety of demographic characteristics and odds of reporting need for support with anxiety, sadness, pain and uncertainty about the future. This work highlights the need to further evaluate the presence of financial distress and toxicity in adults living with CF, as well as to develop interventions to optimise supportive care, accounting for patient demographics.