A combination of quantitative and qualitative methods in investigating risk factors for lost to follow-up for tuberculosis treatment in Japan - Are physicians and nurses at a particular risk?

BACKGROUND

The treatment success rate of pulmonary tuberculosis (PTB) patients aged 64 years and below in Japan, a tuberculosis (TB) middle-burden country with a notification of 13.9 per 100,000 populations in 2016, has been fluctuating around 70% for some years. In order to improve treatment outcome, it is critical to address those lost to follow-up (LTFU). The objective of the study therefore was to describe the characteristics of, and analyze the risk factors for those LTFU among pulmonary TB patients aged between 15 and 64, and discuss policy implications.

METHODS

The study used a mixed method of quantitative and qualitative approach, and was conducted in two phases. The first involved analysis of cohort data from the national TB surveillance of PTB patients newly notified between 1 January 2006 and 31 December 2015. The second phase involved focus group (FGD) discussions with public health nurses, who are responsible for supporting TB patients' adherence to medication, on the possible reasons why some patients become lost to follow-up.

RESULTS

Analysis of the surveillance data suggested that among all patients, positive sputum smear (adjusted odds ratio, [aOR] 0.52, 95% confidence interval [CI] 0.47-0.58) and cavitary lesion on chest x-ray (aOR 0.79, 95%CI 0.72-0.85) decreased the risk, while not requiring hospitalization increased the risk of LTFU (aOR 1.46, 95%CI 1.33-1.60). Among females, being a physician (aOR 2.07 95%CI 1.23-3.48) and nurse (aOR 1.18, 95%CI 1.91-1.37) were identified as additional risk factors for LTFU. The analysis of focus group discussions revealed three possible themes which may be useful in understanding why nurses and physicians were at a higher risk of becoming LTFU-firstly, the possibility that physicians and nurses were finding it difficult to make medication taking a routine, secondly, their low risk perception towards TB is affecting their adherence behavior, and thirdly, their unwillingness to accept DOTS was increasing their risk of becoming LTFU.

CONCLUSIONS

The analysis of surveillance data and FGD transcripts indicated that patient education for those starting their treatment as an outpatient, and establishing DOTS that is both acceptable and realistic to physicians and nurses, may be two issues which need to be addressed urgently.