Impact of a multidisciplinary interstitial lung disease clinic on healthcare utilization

Abstract

RATIONALE

A minimal amount is known of the impact that a multidisciplinary interstitial lung disease (ILD) clinic with dedicated nursing support may have on healthcare utilization.

OBJECTIVES

The objective of this study is to determine if there is a reduction in healthcare utilization including emergency room (ER) visits, hospitalizations, and hospitalization costs in the year after a patient is assessed in the ILD clinic.

METHODS

This retrospective study evaluated the number of ER visits and hospitalizations 1 year before and after being seen in an ILD clinic. For those with hospitalizations, length of stay (LOS) and cost of each stay were collected. Pre- and post-ILD clinic outcomes were compared using Mann-Whitney and Wilcoxon rank sum test.

MEASUREMENTS AND MAIN RESULTS

A total of 202 patients were screened and 140 included in the analysis. Mean age was 66 years (±12), and 47% were female. Mean forced vital capacity percentage predicted was 77% (±22). There was no significant difference in the pre- and post-ILD clinic mean number of ER visits per patient (p = 0.33) nor hospitalizations (p = 0.91). LOS was shorter in the post-ILD clinic period (10.8 ± 10.5 days) versus pre-ILD clinic visit (18.8 ± 25.4 days), although not statistically significant (p = 0.30). Pre-ILD clinic visit mean hospitalization cost was $24,881.89 (±35,817.48) and post-ILD clinic visit was $16,751.68 (±16,549.74) although not statistically different (p = 0.66).

CONCLUSIONS

We found no statistically significant difference in ER visits, hospitalizations or hospitalization costs post-ILD multidisciplinary clinic assessment in our ILD cohort. However, there was a trend toward lower overall healthcare utilization and cost in the year post-ILD clinic.

RÉSUMÉ

JUSTIFICATION:

On sait peu de choses sur les répercussions qu’une clinique multidisciplinaire de pneumopathie interstitielle comptant un soutien infirmier dédié peut avoir sur l’utilisation des soins de santé.

OBJECTIFS:

Déterminer s’il y a une réduction de l’utilisation des soins de santé, y compris les visites aux urgences, les hospitalisations et les coûts d’hospitalisation, dans l’année suivant l’évaluation d’un patient à la clinique multidisciplinaire de pneumopathie interstitielle.

MÉTHODES:

Cette étude rétrospective a évalué le nombre de visites aux urgences et d’hospitalisations un an avant et après qu’un patient ait été vu dans une clinique multidisciplinaire de pneumopathie interstitielle. Pour les personnes hospitalisées, la durée du séjour et le coût de chaque séjour ont été recueillis. Les résultats cliniques avant et après avoir été vu à la clinique multidisciplinaire de pneumopathie interstitielle ont été comparés à l’aide du test de somme des rangs de Mann-Whitney et Wilcoxon.

MESURES ET PRINCIPAUX RÉSULTATS:

202 patients ont été sélectionnés et 140 ont été inclus dans l’analyse. L’âge moyen était de 66 ans (±12 ans), et 47 % étaient des femmes. Le pourcentage moyen de capacité vitale forcée prévu était de 77 % (±22). Il n’y avait pas de différence significative dans le nombre moyen de visites aux urgences par patient avant et après la clinique multidisciplinaire de pneumopathie interstitielle (p = 0,33) ni d’hospitalisations (p = 0,91). La durée du séjour était plus courte dans la période clinique après la clinique multidisciplinaire de pneumopathie interstitielle (10,8 ± 10,5 jours) par rapport à la visite clinique avant la clinique multidisciplinaire de pneumopathie interstitielle (18,8 ± 25,4 jours), bien que non statistiquement significatif (p = 0,30). Le coût moyen d’hospitalisation avant la visite à la clinique multidisciplinaire de pneumopathie interstitielle était de 24 881,89 $(± 35 817,48) et le coût moyen d’hospitalisation après la visite était de 16 751,68 $(±16 549,74), bien qu’il ne soit pas statistiquement différent (p = 0,66).

CONCLUSIONS

Nous n’avons trouvé aucune différence statistiquement significative dans les visites aux urgences, les hospitalisations ou les coûts d’hospitalisation après l’évaluation par une clinique multidisciplinaire de pneumopathie interstitielle dans notre cohorte. Cependant, une tendance à la baisse de l’utilisation et du coût global des soins de santé au cours de l’année suivant la clinique multidisciplinaire de pneumopathie interstitielle a été observée.

Keywords:

• Healthcare utilization
• interstitial lung disease
• multidisciplinary care
• outcomes
• pulmonary fibrosis

Introduction

Interstitial lung diseases (ILD) are a group of more than 200 specific lung diseases that cause inflammation and/or scarring of the lung and includes idiopathic pulmonary fibrosis (IPF).^1–3 IPF is a chronic, progressive and scarring disease limited to the lung with few treatment options and a median survival time of 3–5 years from diagnosis without treatment.⁴ In Canada, the estimated prevalence of IPF in 2011 was just over 14,000 with an incidence of 18.7 per 100,000 persons.⁵

A multi-year study in Quebec, Canada found that annual health care costs for patients with IPF alone begin to increase prior to diagnosis.⁶ In Quebec, the average annual cost per patient 2 years prior to IPF diagnosis was $7,049 Canadian dollars (CAD), rising to $11,664 CAD 1 year prior. The average annual cost per patient was $17,398 CAD the year after diagnosis, excluding the cost of the original diagnosis (index admission) of $15,281 CAD. The average cost remained above $12,000 CAD for the second-year post-diagnosis. Another study in Alberta, Canada evaluated the differences in resource use and associated costs of end-of-life care between patients with IPF receiving early integrated palliative care with nursing support and found that their intervention resulted in improved quality of end-of-life care and reduction in costs.⁷ In the United States (US) Medicare population, patients with IPF incurred substantial health care resource utilization, with an annual IPF-attributable medical cost to the US Medicare system, excluding medication costs, estimated at close to $2 billion.⁸ A total of 7855 patients with IPF were matched to 38,856 controls. Compared with matched controls in the year before diagnosis, patients with IPF had an 82% higher risk of hospitalization (28.8 vs. 15.8%), and 72% higher total medical costs ($10,124 vs. $5,888). Compared with matched controls during a 1-year period after diagnosis, patients with IPF had a 134% higher risk of hospitalization (48.7 vs. 20.8%), similar increased risk of emergency room (ER) visits (39.6 vs. 17.5%), and 134% higher total medical costs ($20,887 vs. $8,932). A more recent study evaluated hospitalizations and costs in ILD in the US from 2008 to 2018 and found that over that time hospital costs were increasing while the length of stay (LOS) was stable.⁹ Patients with progressive fibrosing ILD when compared to those who did not progress also demonstrate significant increases in healthcare utilization and cost over a 12-month period in the US.¹⁰ These are significant costs to healthcare systems and healthcare utilization is an area for potential improvement by enhancing outpatient support with integrated care delivery including specialized nursing. Specialized nursing has been shown to have a positive impact on care delivery including cost-effectiveness in outpatient roles, including heart failure, diabetes and chronic obstructive pulmonary disease management.^11–15 ILD nurses can support clinical care in multiple ways, such as disease education, counseling on medications and oxygen, assistance in managing medication side effects, communication in times of acute change between patient and physician and caregiver support. We hypothesized that after contact with an ILD clinic that includes a dedicated nurse for patient support and care, healthcare utilization would be reduced at our center.

Materials and methods

Study population

All patients seen in the Saskatchewan ILD clinic starting from January 2018 when this clinic was initiated were assessed for eligibility in this retrospective study. Once seen in this clinic patients were met by and had access to the ILD nurse including for medication and phone support which could be triggered by the nurse, patient or physician. This study received ethics approval from the University of Saskatchewan (Bio-ID 2070). Inclusion criteria included age ≥18 and at least 12 months of follow up prior to March 2019 to reduce potential bias of those who avoided hospital contact during the start of the SARS-CoV-2 coronavirus pandemic starting in March 2020. Patients were excluded if they were not followed in the ILD clinic after the initial visit. Baseline variables collected at the first ILD clinic visit included age, patient reported sex, smoking status (current/ex-smoker versus never smoker), forced vital capacity (FVC) absolute and percent predicted, diffusion capacity of the lung (DLCO) absolute and percent predicted, and diagnosis. Diagnosis was ascertained both through multidisciplinary discussion (MDD) and in some cases by the clinicians alone in combined clinics run with rheumatology.

Healthcare utilization

Data was sourced from the Discharge Abstract Database and National Ambulatory Care Reporting System and included ER registration dates, admission and discharge dates for hospitalizations and the most responsible diagnosis for hospitals in both major cities in Saskatchewan, Canada - Regina and Saskatoon. Data for centers outside these cities were not available at the time of this study. All ER visits and hospitalization in these centers were counted within the year prior to being seen and one year after being seen in the ILD clinic. Reasons for ER visits and hospitalizations were broken down into respiratory, surgical lung biopsy, cardiac, neoplasm, convalescence, urinary tract, renal, infection, neurologic, stem cell transplant, therapy, gastrointestinal, orthopedic and palliative, among others. Length of stay (LOS) was also recorded for each hospitalization and calculated as the date of discharge minus the date of admission plus 1 (to include all days in the estimate). The resource intensity weight (RIW) for each hospitalization was also recorded. RIWs were obtained from the Canadian Institute for Health Information and reflect the intensity of resources consumed during a hospital admission for a group of similar conditions and consequently adjusts the estimated cost associated with that hospital admission.^16,17

Hospitalization costs

Costs of hospitalization were calculated using case mix grouper methodology.¹⁸ Canadian Institute for Health Information was accessed for Saskatchewan average hospital cost for the year of the hospitalization and multiplied by the RIW to obtain the estimated total cost per hospitalization. This was then adjusted for inflation to reflect a cost for April 2021. Proximity to the Regina and Saskatoon hospitals was also recorded per patient (within or beyond 70 kilometers [km]) as smaller hospitals may have been the closer facility beyond this distance.

Statistical analysis

Descriptive statistics were performed to summarize the data. Mean, standard deviation estimates (±SD), median (interquartile range [IQR], and the range of ER and hospitalizations, LOS and cost were calculated separately and compared between pre- and post-ILD clinic visit time periods using Mann-Whitney U tests. A sub-group analysis comparing outcomes among those within 70 km of a tertiary care center with those more than 70 km outside a tertiary care center was also performed. To investigate the differences between pre- and post-ILD clinic visits, binary outcomes were created for ER visits and hospital visits since the number of ER and hospital visits for this data set was dispersed with 85% having zero visits. Multivariate logistic regression models for binary outcomes ER and Hospital visits were fitted and adjusted for age, sex and baseline FVC%. Statistical analyses were performed with the IBM SPSS Statistics Version 28 and Stata/SE 17.0. Statistical significance was set by an alpha level 0.05.

Results

Patient characteristics

A total of 202 patients were consecutively screened from January 2018 to March 2019 and 140 met the inclusion criteria. Forty-seven patients were excluded due to having less than 12 months of follow-up data and 15 patients due to not being followed by the ILD clinic after the first visit. Baseline characteristics are presented in Table 1. The average age was 66 years (±12), and 47% were female. Mean FVC% was 77% (±22) and mean DLCO% was 53% (±19). Treatment was started or changed after the initial visit in 63% of patients. During the 1-year follow up period there were 4 deaths and 1 transplant. The predominant diagnosis was connective tissue disease-associated ILD (CTD-ILD)/interstitial pneumonia with autoimmune features (IPAF) (41%) followed by fibrotic hypersensitivity pneumonitis (23%), IPF (16%) and unclassifiable (9%) (Figure 1). The referral base breakdown was as follows: respirologist 63%, family physicians or primary care providers 20%, rheumatology 8%, general medicine or intensive care 5%, cardiology and nephrology 4% and thoracic surgery 0.7%.

Figure 1. Distribution of interstitial lung disease diagnoses in the clinic. Other = asbestosis (1), DIP (2), drug reaction (1), PBC ILD (2), RB ILD (2). Abbreviations: ILD, interstitial lung disease; CTD ILD, connective tissue disease associated ILD; IPF, idiopathic pulmonary fibrosis; HP, hypersensitivity pneumonitis; DIP, desquamative interstitial pneumonia; PBC, primary biliary cirrhosis; RB, respiratory bronchiolitis.

Table 1. Patient characteristics.

Variable	n = 140
Age (years), mean (±SD)	66 (±12)
Female sex, n (%)	66 (47)
Current or ex-smoker, n (%)	91 (65)
Median pack year history (IQR)	20 (25)
FVC L, mean (SD)	2.9 (±1.0)
FVC%, mean (SD)	77 (±22)
DLCO ml/mmHg/min, mean (SD)	12.8 (±5.2)
DLCO%, mean (SD)	53 (±19)
Distance within 70 km (home to tertiary hospital), n (%)	65 (46)
Deceased, n (%)	4 (3)
Transplant, n (%)	1 (0.7)
Treatment initiated/changed after being seen at ILD clinic, n (%)	88 (63)

Abbreviations: n: number; SD: standard deviation; IQR: interquartile range; FVC: forced vital capacity; DLCO: diffusing capacity of the lung.

Emergency room visits

For the full cohort in the 1-year leading up to the initial ILD clinic visit, there were a total of 46 ER visits, 13 (28%) of which were respiratory related. Twenty-one patients accounted for these visits with 12 patients having visited once and one patient with 9 ER visits. In the first year after the ILD clinic visit, the total number of ER visits was 66, 6 (9%) of which were respiratory related. Twenty-seven patients accounted for these visits with 15 patients having visited once and one patient with 12 ER visits (Figure 2a). The mean number of ER visits per patient in the pre-ILD clinic period was 0.3 (±1.1) and for the post-ILD clinic period was 0.5 (±1.5) with no statistical difference between the two (p = 0.33). The odds of an ER visit were not significantly different in the post-ILD clinic visit period in both the unadjusted (p = 0.27) and adjusted models (p = 0.28). For those who lived within 70 km of a tertiary care center the mean numbers of ER visits per patient in the pre-ILD clinic period was 0.7 (±1.6) and was 1.0 (±2.1) in the post-ILD clinic visit which was not significantly different (p = 0.17).

Figure 2. Pre- and post-ILD clinic visit comparisons in number of ER visits (a), hospitalizations (b), mean LOS for hospitalizations (c), and mean hospitalization costs (d) for both all indications (total) and for respiratory indications only. Abbreviations: ILD, interstitial lung disease; ER, emergency room; LOS, length of stay.

Hospitalization

For the full cohort in the year leading up to being seen in the ILD clinic, there were 48 hospitalizations from a total of 27 patients, with 18 of these patients having had one admission each and the remainder were multiple admissions. Of these 48 admissions, 24 (50%) were for a respiratory indication. In the year after being seen in the ILD clinic there were 42 hospitalizations from a total of 28 patients, with 20 of those patients having had one admission each and the remainder were multiple admission. Of these 42 admissions, 18 (43%) were for a respiratory indication (Figure 2b).

The mean number of hospitalizations in the pre-ILD clinic period was 0.3 (±1.0) and for the post-ILD clinic period was 0.3 (±0.7) and was not significantly different (p = 0.91). The odds of a hospitalization were not significantly different in the post-ILD clinic visit period in both the unadjusted (p = 0.87) and adjusted models (p > 0.99). For those who lived within 70 km of a tertiary care center the mean numbers of hospitalizations per patient in the pre-ILD clinic period was 0.6 (±1.4) and was 0.4 (±0.8) in the post-ILD clinic visit which was not significantly different (p = 0.52).

Hospitalization length of stay

The mean LOS for hospitalizations in the year leading up to the initial ILD clinic visit was 18.8 (±25.4) days, and for respiratory-related admissions the LOS mean was 12.0 (±16.7) days. In the year after the initial ILD clinic visit, the mean LOS was lower at 10.8 (±10.5) days, and for respiratory related admission the LOS mean was also lower at 7.6 (±5.6) days (Figure 2c). There was no statistically significant difference in mean LOS in the period before and after the ILD clinic for the full cohort (p = 0.30) nor for those within 70 km (p = 0.17).

Hospitalization cost

For the hospitalizations that occurred in the year prior to being seen in the ILD clinic, the total hospitalization cost was $671,810.97, and for the respiratory related visits the total cost was $335,870.51. Mean cost per visit was $24,881.89 (±35,817.48) for all visit types, and for the respiratory-related visits was $21,505.07 (±35,204.86). For the hospitalizations that occurred in the year after being seen in the ILD clinic, total hospitalization cost was lower at $398,204.65, and for respiratory related visits the total cost was also lower at $225,929.22 (Figure 2d). Mean cost per visit was $16,751.68 (±16,549.74) for all visit types, and for respiratory-related visits was $12,973.48 (±11,412.61), both lower than pre-ILD clinic hospitalizations. The difference in hospitalization cost for all visit types was not statistically different (p = 0.66).

Discussion

This retrospective study evaluated healthcare utilization among a well-defined cohort of patients with ILD in the year prior to and after being assessed in an ILD clinic with nursing support. We found that after being seen in an ILD clinic, there is a numerical reduction in the number of respiratory-related ER visits and hospitalizations (all indications and respiratory-related indications), although the results were not statistically significant. There was also a numerical reduction in the overall LOS in the post-ILD clinic period. Of significant interest is the reduction in overall costs related to the hospitalizations with a difference of $273,606.32 in overall hospitalization cost between the two periods, as well as when looking at the respiratory-related hospitalizations with a difference of $109,941.29. Mean cost per visit also was lower by $8,130.21 for all visit types, and by $8,531.59 for respiratory-related visits.

There are several limitations to this study including the retrospective nature and small patient and event numbers. We were only able to evaluate ER visits and hospitalizations in the tertiary care hospitals during this time period given limitations with data availability. However, we did evaluate those who live within 70 km of the centers and still found no significant difference in this group. Hospitalizations and ER visits are also not independent of each other, and it is difficult to control for this factor alone. Additionally, we recognize that costs associated with healthcare in general are complex, and we do not account for outpatient diagnostics, treatment costs, or cost in general of ongoing care in this study. Regardless, it is difficult to ignore the vast difference in hospitalization costs after being seen in the ILD clinic and although the LOS was not statistically significant, there is a numerical difference that in times of increasing strain on the healthcare system should not be disregarded. We are not able to identify exact interventions that may have led to this difference, however being assessed in the ILD clinic often led to new treatment applications that may favorably alter a patient’s overall health which may also impact the need to seek out emergent healthcare. However, they also gain direct access to support from the ILD nurse. ILD nursing support is multifactorial including phone advice, communication with the ILD physician and early intervention for symptoms that may signal issues with medications, new respiratory or cardiac symptoms and subsequent timely outpatient care. Additionally, given that patients were referred by other respirologist in 63% of the cases, this may select for a more advanced ILD population that may lead to more ER visits and hospitalizations.

In conclusion, we found that in the year after being seen in an ILD clinic, patients utilized ER and hospital care less frequently for respiratory-related events, had less hospitalizations and length of stay and the cost of hospitalizations was also reduced. Given our small numbers it was not possible to demonstrate statistically significant results; however, these numerical differences are of considerable interest. Multidisciplinary care of ILD patients leads to improved overall care and reduced the cost of tertiary hospital care which indicates a meaningful investment for healthcare systems. Future study evaluating overall cost-effectiveness factoring in all relevant costs including nursing salary, medication costs, quality of life, survival and quality-adjusted life-year would be of great interest for all stakeholders.

Author contributions

V. Marcoux was involved in conception and design, data analysis and interpretation. H. Le and P. Mondal were involved in data analysis and interpretation. J. Marciniuk, P. Landry and E. Penza were involved in data analysis and interpretation. All authors were involved in manuscript writing including final approval.

Acknowledgments

The authors would like to thank the dedicated ILD nurses who work tirelessly in this field to improve the lives of those with these diseases and the patients for their valuable contribution to research.

Disclosure statement

V. Marcoux reports personal fees from Boehringer Ingelheim Canada and Hoffman-La Roche Ltd, and Astra Zeneca, grants from the University of Saskatchewan, Royal University Hospital Foundation, Boehringer Ingelheim, Astra Zeneca and Hoffman-La Roche. J. Marciniuk reports personal fees from AstraZeneca. P. Landry reports personal fees from Glaxo Smith Kline, Boehringer Ingelheim Canada, AstraZeneca, Valeo, MEDUCOM and LiiV agencies. E. Penza reports personal fees from AstraZeneca, GlaxoSmithKline, Sanofi, Covis, LungSask, Canadian Thoracic Society; research funds from CIHR, Saskatchewan Health Research Foundation, Respiratory Research Center, LungSask, Saskatchewan Center for Patient oriented research, and AstraZeneca. She reports contract with Saskatchewan Cancer Agency for her role as medical advisor for the Saskatchewan Lung Screening program, board member of Institute of Cancer Research Advisory Board (CIHR), Canadian Thoracic Society Executive Committee, and Past Co-Chair of the CTS COPD Clinical Assembly. H. Le and P. Mondal have no disclosures.

Data availability statement

The data for this data is safeguarded by VM and used with the approval of the University of Saskatchewan ethics board and are not available.

Additional information

Funding

The author(s) reported there is no funding associated with the work featured in this article.