A total of 5,628 patients with NTM-PD were identified in the SNDS database between 2010 and 2017, 4,898 patients from hospitalizations and 730 from outpatient drug consumptions. Among patients identified through hospitalizations, 703 had a specific treatment for NTM-PD vs 4,195 with no treatment. The mean age was 60.9 years (SD ± 19.5) and 52.9% were males. The Universal Health Coverage for low-income individuals was observed in 8.8% of patients (Table 1).
A total of 3,954 patients were diagnosed with NTM-PD between January 01/2010 and December 31/2017; thus, the prevalence of NTM-PD was estimated at 5.92 per 100,000 inhabitants over 8 years. The incidence rate of NTM-PD (Figure 1) remained stable over time, with a min of 1.025/100,000 in 2010 (N=662) and a max of 1.096/100,000 (N=732) in 2017 with slight variations in-between years.
Matched controls could not be identified for all patients in the study; therefore, comparisons included 4,447 NTM-PD cases matched to 13,341 controls for age, gender, and region.
Analysis of comorbidities (Table 1) found a significantly higher Charlson Comorbidity Index (CCI) mean score in patients with NTM-PD (1.6) compared to controls (0.2) (p<0.0001). However, there was no difference between treated and untreated groups on the CCI score (p=0.3). Overall, patients with NTM-PD had a higher proportion of risk factors compared to controls including respectively corticoid treatment in the last 3 years (57.3% versus 33.8%), chronic lower respiratory disease (34.4% vs 2.7%), other infectious pneumonia (24.4% vs. 1.4%), malnutrition (22.0% vs. 2.0%), and tuberculosis (14.1% vs. 0.1%).
Even if statistically significant, no notable differences between the treated and untreated group were found, except higher rates of HIV infection in treated vs. untreated patients (14.8% vs 5%).
A first analysis showed that the mortality of 4,447 NTM-PD cases was significantly higher than that of 13,341 controls (p<0.0001) (Figure 2).
In multivariate analysis (Table 2), after adjustment for age, residence, gender, residence, and risk factors, the risk of mortality was more than 2 times higher in NTM-PD patients compared to controls (HR=2.8 (95% CI [2.5; 3.1]). The risk of mortality was more likely to increase in older patients with NTM-PD including those aged 70-79 years (HR=16.081, 95% CI: 6.646-38.908), 80-89 years (HR=33.654, 95% CI: 13.910-81.423), and more than 90 years (HR=56.860, 95 CI: 23.209-139.303). Mortality was also higher in patients with Universal Health Coverage, HIV infection, lung cancer and graft, other infectious pneumonia, malnutrition, and other chronic obstructive pulmonary disease. Mortality was lower in female patients vs. males and in patients who had mucopurulent chronic bronchitis and bronchiectasis.
In a second multivariate analysis performed on the NTM-PD cohort and adjusted for age, residence, gender, residence, and comorbidities, the risk of mortality was lower for treated patients with antibiotics compared to untreated patients (HR=0.772 (95% CI [0.628; 0.949]) (Table 3).
Out of the 5,628 patients, 25.5% (1,433) had received antibiotics to treat NTM-PD vs 74.5% (4,195) with no treatment (Table 4).
Among the 4,898 hospitalized patients, 703 (14.4%) had a specific treatment for NTM and 4,195 (85.6%) did not receive any treatment. The most frequently used treatment first line regimens were: Clarithromycin+Ethambutol for 34.4% of patients, Clarithromycin + Rifampin+ Ethambutol (22.1%), Clarithromycin+Rifampin (10.3%), and Clarithromycin monotherapy (9.3%). Other combinations represented less than 5% of patients for each of them (Additional table A5).
Figure 3 depicts the proportion of patients maintaining treatments at 3, 6, 9, and 12 months. Half of the patients (56%) were still treated at 3 months, 40% at 6 months, 30% at 9 months, and only 22% at 12 months. The majority of patients discontinued before 12 months of treatment.
This analysis included 3,950 cases and 11,850 controls. A total of 3,641 NTM-PD patients (92.2%) had been hospitalized at least once since diagnosis compared to 2,484 controls (21.0%). For those hospitalized at least once, the length of hospital stay was 40.3 days for cases vs 16.5 days for controls; mean hospital stay per patient was 5.0 for cases and 2.7 for controls.
A total of 2,683 cases and 8,049 controls were included in the economic analysis.
The mean total cost reimbursed in the year following NTM-PD diagnosis was significantly higher for cases compared to controls, €22,966 vs €2,709, respectively (p<0.0001). Total expenses (societal perspective) followed the same trend with a greater amount for cases than controls, €24,083 versus €3,402, respectively (p<0.0001).
From the French Public Health Insurance perspective (Figure 4) the mean cost for hospitalizations in the year following NTM-PD diagnosis was €12,524 for cases versus €1,156 for controls (p<0.0001). The total cost for outpatient care in the year following NTM-PD diagnosis was €10,442 for NTM-PD patients and €1,553 for controls. Costs of drugs were estimated at €5,493 for cases (52.6% of the total amount) versus €517 for controls (33% of the total amount).
From a perspective incl. out-of-pocket costs (Figure 4), the mean cost for hospitalizations in the year following NTM-PD diagnosis was €12,499 for cases versus €1,303 for controls (p<0.0001). The total cost of outpatient care in the year following NTM-PD diagnosis was €11,584 for NTM-PD patients and €2,099 for controls. Costs of drugs were estimated at €5,872 for cases (50.6% of the total amount) versus € 629 for controls (30% of the total amount).