Delayed Tuberculosis Diagnosis in a Low Incidence Country and its Effect on Cost of Care


 Background

Tuberculosis elimination (TB) is a global priority that requires high-quality timely care to be achieved. In low TB incidence countries such as Ireland, delayed diagnosis is common. Despite cost being central to policy making, it is not known if delayed care affects care cost among TB patients in a low-incidence setting.
Methods

Health care records of patients with signs and symptoms of TB evaluated by a tertiary service in Ireland between July 1st 2018 and December 31st 2019 were reviewed to measure and determine predictors of patient-related delays, health care-provider related delay and the cost of TB care. Benchmarks against which the outcomes were compared were derived from the literature.
Results

Thirty-seven patients were diagnosed with TB and 51% (19/37) had pulmonary TB (PTB). The median patient-related delay was 60 days among those with PTB, greater than the benchmark derived from the literature (38 days). The median health care provider-related delay among patients with PTB was 16 days and, although similar to the benchmark (median 22 days, minimum 11 days, maximum 36 days) could be improved. The health care-provider related delay among patients with EPTB was 66 days, greater than the benchmark (42 days). The cost of care was €8298, and while similar to that reported in the literature (median €9,319, minimum €6,486, maximum €14,750) could be improved. Patient-related delay among those with PTB predicted care costs.
Conclusion

Patient-related and health care-related delays in TB diagnosis in Ireland must be reduced. Initiatives to do so should be resourced.


Introduction
In 2020, 10 million people were infected with tuberculosis (TB), a preventable infectious disease [1]. Despite being treatable, 1.3 million people died with TB in 2020. TB elimination (an incidence less than 1 case per million [2]) is a global priority [3] [4]. Internationally, there is a growing consensus that to achieve this, a paradigm shift from a focus on only TB care coverage to one that includes care quality is needed [1][2] [5]. High-quality TB care includes timely diagnosis and treatment initiation. Delays in TB treatment are associated with more severe disease, greater mortality, and a risk of ongoing transmission [6,7][8]. Therefore, TB programmes must be able to identify not only people atrisk of TB disease but those most at-risk of delayed diagnosis and treatment. However, due to the low visibility of TB in countries with a low incidence, programmatic activities to enable timely diagnosis and treatment are often not prioritised by health care managers or policy makers for resourcing [2]. Delays in the diagnosis and treatment of TB have been reported in Ireland [9] [10]. In countries with a low incidence of TB, such as Ireland, how delayed TB care effects cost of care is not known despite cost being central to many policy decisions. The primary aim of this evaluation was to determine factors that predict patient-related and health care provider-related delays in TB diagnosis and treatment and to establish if the cost of TB care in a low-incidence country is affected by delays in care.

Methods
A retrospective review of the health care records of patients evaluated by a tertiary centre TB service in Ireland was performed. Patients who were referred with symptoms or signs of active TB between the 1st of July 2018 and the 31st of December 2019 were included. Patients with non-tuberculosis mycobacteria infection were excluded. Data extraction was performed by author 1 and author 2, with author 1 reviewing all data collected for accuracy. Data collected included age, sex, nationality and TB disease site, microbiological characteristics (culture, GeneXpert and smear status on presentation), human immunode ciency virus (HIV) status, place of rst presentation (primary care, outpatient clinic, emergency department), and resource utilisation (Appendix 1). Patient-related delays were de ned as the time from symptom onset to rst presentation to health care services. Health care-related delays were de ned as the time from rst presentation to health care services to treatment initiation. Pulmonary TB (PTB) cases were de ned as any patient with involvement of the lung parenchyma or the tracheobronchial tree [11]. Extra-pulmonary TB cases (EPTB) were those with TB not captured by the de nition of PTB. Treatment outcomes were reported according to WHO de nitions [11].
Cost calculations considered direct costs, those that relate to TB diagnosis and treatment. These included the cost of radiological (radiographs, computed tomography imaging, magnetic resonance imaging, ultrasonography, positron-emission tomography), haematological (full blood counts), biochemical (renal and liver function pro les), immunological (interferon gamma release assay), virological (hepatitis B, hepatitis C, HIV testing) and microbiological (TB smear and culture, GeneXpert testing) investigations. Investigation costs (Appendix 1) were sourced primarily from the laboratory directorate, literature published from Ireland, and for radiological tests, the National Health Service cost collection [12]. Anti-tuberculosis medications and pyridoxine reimbursement costs were sourced from the Health Service Executive (HSE) (Appendix 1). Where drug costs were unavailable from the HSE, costs were sourced rst from MIMS Ireland or second from the British National Formulary. The drug costs to the health system were calculated following guidance from the National Centre for Pharmacoeconomics [13]. Cost data were in ated to 2019 values and costs in British Pounds Sterling were converted to Euro using Organization for Economic Co-operation and Development (OECD) purchasing power parity gures following national guidance from the Health Information and Quality Authority [14]. The cost of outpatient TB and ophthalmology clinics were calculated according to national costing guidelines and HSE salary scales [15][16](Appendix 1). Hospitalisations due to TB, either known or unknown at the time of admission, were costed following guidance from the Healthcare Pricing O ce [17] and included elective, emergency and rehabilitation related hospitalisations (Appendix 1).
A structured review of the literature was performed to identify studies which reported patient-related delays, health care provider-related delays and direct costs of TB care (Appendix 2). Outcomes from this evaluation were benchmarked against those reported in the literature from other low incidence countries to determine if care was delayed or more costly.
The median and interquartile range of patient related-delays, health care provider-related delays and direct costs of care for patients with TB were reported. Analyses were conducted separately for patients with PTB and EPTB. Multivariable regression was performed with patientrelated delay as the dependent variable and age, sex, nationality, and HIV status as the independent variables. Multivariable regression was performed with health care provider-related delay as the dependent variable and age, sex, nationality, HIV status, culture status, drug resistance status and acute hospitalisation requirement as the independent variables. Multivariable regression was also performed with direct cost of care as the dependent variable and age, sex, nationality, HIV status, culture status, patient-related delays, and health careprovider related delays as the independent variables. The median, minimum and maximum reported values in studies retrieved from the literature review for patient-related delays, health care provider-related delays and cost of drug-susceptible TB care were reported. Costs reported in the literature were in ated to 2019 values using Organization for Economic Co-operation and Development (OECD) consumer price index data [18] and then converted to Euros using the purchasing power parity index as reported by the OECD [19].

Results
Fifty-four patients were assessed for TB during the reference period. Most patients (69% (37/54)) referred had a diagnosis of TB made. Over half (51% (19/37)) of patients with TB had PTB (Table 1). Males comprised 57% (21/37) of patients. The median age of TB patients was 41.7 years (Interquartile range 35.1-46.3). Sixty per cent of patients (22/37) were from a country with a high incidence of TB (≥40 cases annually per 100,000 of population). Most patients (87% (32/37)) reviewed had a risk factor for TB. While screening for HIV in the TB clinic was imperfect, only 8% (3/37) of TB patients had a diagnosis of HIV. No TB patient had a history of treatment for LTBI.
Among patients who remained under the care of the TB service, hospitalization occurred in 91% (32/35) of patients and amounted to 1484.5 bed-days. Over half (51% (18/35)) of all TB patients had an emergency hospitalization. Illness due to undiagnosed TB requiring management or investigation was the reason for 86% (44/51) of hospitalizations and the remainder of admissions related to rehabilitation in patients treated for TB (6% (3/51)) or management of disease or treatment-related complications (8% (4/51)).

Patient-related delays
Forty-six percent (17/37) of patients presented rst to their primary care physician with their symptoms prior to diagnosis, 41% (15/37) presented with their symptoms directly to an emergency department and 13% (5/37) patients presented their symptoms to a physician while attending another hospital outpatient clinic. Overall, 59% (22/37) of patients had their diagnosis established by attending the emergency department.
In patients with PTB, the median patient-related delay was 60 days (IQR 30-180). In a multivariable regression model (   730-18,057). In a multivariable model that included age, sex, nationality, HIV status, culture status, patient-related delays, and health care provider-related delays only patient-related delays predicted a higher cost of care among patients with PTB who completed treatment (Table 3). In a multivariable model that included age, sex, nationality, HIV status, culture status, patient-related delay, and health care provider-related delay, no variable predicted the cost of TB care among patients with EPTB.  [20]. The median, minimum and maximum health care-related delays among patients with PTB are 22 days, 11 days, and 36 days, respectively. The median health care provider related delay among patients with PTB in this evaluation of 16 days is similar to that reported in the literature but suggests it could be improved.
With regard to patients with EPTB, there was signi cant variation in the patient-related delay reported in the literature among differing EPTB disease sites and symptom complexes. Therefore, a benchmark for EPTB patient-related delay could not be reliably de ned for comparison with the outcomes reported in this evaluation. Among patients with EPTB, the health care-related delays reported in the included studies were 32, 39 and 42 days [22][28], all lesser than the health care-related delay among patients with EPTB in this evaluation of 66 days.
Regarding the direct cost of TB care, the median, minimum and maximum costs of non-rifampicin resistant non-multidrug resistant TB care reported in the included studies were €9,319, €6,486, and €14,750, respectively (Table 4). Therefore, the cost of care reported in this evaluation (€8298), while similar to that in other low incidence countries, suggests it could be improved.

Discussion
This evaluation described the timeliness and cost of TB care in a low incidence country, determined predictors of delayed care and, for patients with PTB, demonstrated that longer patient-related delays resulted in increased care costs. Patient-related delays were substantial, particularly among patients with PTB. Although patient-related delays for those with EPTB were not compared with outcomes reported in the literature, the absolute value reported in this study of 30 days was long. Patient characteristics such as age, sex, nationality, or HIV status did not predict patient-related delays among those with PTB. HIV infection predicted patient-related delays among those with EPTB. This highlights a need to ensure HIV is diagnosed early and that people living with HIV are kept engaged in HIV care services that can diagnose TB and provide patients with education about its signs and symptoms. Other factors, such as the absence of universal health care in Ireland, may have contributed to the long patient-related delays reported in this evaluation. Cost and long-waiting times have been reported as barriers to accessing health care in Ireland [37][38], particularly in vulnerable groups who may face additional challenges to accessing health care services. In this evaluation, only 46% of patients presented rst to their primary care physician supporting the assertion that the absence of universal health care may have been a factor. Health care provider-related delays for patients with TB were prolonged, particularly for patients with EPTB (66 days). Culture positivity predicted a shorter health care provider-related delay among those with PTB, demonstrating the importance of performing culture of respiratory specimens for TB early in symptomatic patients. Increasing patient-related delays in those with PTB resulted in increased care costs, with each day of delay increasing costs by €424. Although the median cost of TB care was similar to that reported in the literature, it could be improved, particularly for PTB, which was more costly than EPTB care in this evaluation.
This study adds to the literature on TB care in low-incidence countries where delays in accessing and diagnosing TB are a well-recognised challenge [23][24] [25], which has also been reported in other studies in Ireland [9] [10]. The association between increasing costs of PTB care and patient-related delays has not previously been reported in the literature from countries with a low incidence of TB.
A limitation of this study was the retrospective means by which data were collected from healthcare records. Additional healthcare resource utilisation in primary care or private health facilities may not have been documented in these records. Therefore, the cost of care reported in this study could be an underestimate of the true value. The single-centre nature of this study may limit the generalisability of the cost estimates. However, hospitalizations were the greatest component of TB care cost (92.1%) and a similar proportion of TB cases being hospitalized has been reported nationally, as in this study [39], supporting the generalisability of the direct cost estimates to TB care in other centres in Ireland. There were no patients with rifampicin or multidrug-resistant TB, meaning the results cannot be generalised to these patient cohorts.
Future research is needed in Ireland to better understand the causes of patient-related and health care provider-related delays in TB diagnosis and treatment, many of which will be country and population speci c. Patient-pathway analysis has been useful in other countries in doing this [40]. Qualitative research that evaluates patients' experiences of seeking care prior to diagnosis should be performed to identify barriers to utilising and accessing health care with their symptoms, particularly in primary care. The national TB programme in Ireland should initiate such research studies to determine how TB care in primary and secondary care services can meet the needs and expectations of patients with TB prior, during, and after their diagnosis. By doing so, appropriate initiatives to reduce patient-related and health care-provider related delay can be implemented.
The ndings of this study have implications for the national TB programme in Ireland, which should be ensuring patients with TB receive high-quality timely and effect care. Researchers, healthcare professionals and medical organisations have regularly highlighted insu cient resourcing for TB services in Ireland, where there is no dedicated funding for the TB program [9]. Unless initiatives to reduce TB care delays are identi ed and resourced, patients with TB in Ireland will continue to have prolonged morbidity, ongoing opportunities for transmission, and incur higher care costs. In this context, TB elimination will likely not be achieved nationally.

Conclusion
Patient-related and health care-related delays in TB diagnosis and treatment in Ireland must be reduced. Initiatives to do so should be resourced nationally and if effective could reduce the cost of PTB care.

Declarations Funding
The salary of the rst author of this evaluation was funded by the Royal College of Surgeons in Ireland.

Con icts of interest/Competing interests
The authors have no con icts of interest to declare.

Availability of data and material
The datasets generated during and/or analysed during the current study are not publicly available because they were collected as part of a quality of care evaluation but are available from the corresponding author on reasonable request.
Code availability Not applicable.

Ethics approval
This study was a quality of care service evaluation and it was registered with the Beaumont Hospital O ce of Clinical Audit (approved audit number 880).
Consent to participate