Abstract
Background
Although tuberculosis (TB) treatment success rates are high globally, growing evidence indicates that many TB survivors experience persistent respiratory symptoms, impaired quality of life and reduced well-being after treatment completion. Delayed TB diagnosis may exacerbate these long-term consequences by prolonging untreated disease and lung injury. However, longitudinal evidence linking diagnosis delay to longer-term outcomes remains limited. This study examines the association between TB diagnosis delay and respiratory health, quality of life and mental health among TB survivors in Cambodia.
Methods
A prospective cohort study of adults with newly diagnosed TB was conducted with a baseline in 2022 and 1 year and 2 year follow-ups. TB diagnosis delay was defined as the time from symptom onset to diagnosis and categorised as short or long using the median (64 days). Outcome variables included respiratory health, quality of life and depressive symptoms, assessed with the St George’s Respiratory Questionnaire for Chronic Obstructive Pulmonary Disease (SGRQ-C), WHO Quality of Life–Brief (WHOQOL-BREF) and Centre for Epidemiologic Studies Depression Scale, 10-item version (CES-D-10), respectively. Linear mixed-effects models, adjusted for potential confounders, were applied.
Results
Among 712 participants at baseline, the long-delay group had worse respiratory health before treatment completion and over 2 years, with higher SGRQ-C total (β=4.37, 95% CI 1.10 to 7.64), symptom (β=5.83, 95% CI 2.34 to 9.31) and activity scores (β=6.95, 95% CI 2.67 to 11.23) than the short-delay group. Long delay was also associated with poorer psychological (β=−3.17, 95% CI −5.31 to −1.03) and environmental domains (β=−1.74, 95% CI −3.07 to −0.4) of the quality of life but not with its physical and social domains or depressive symptoms. The differences between the groups remained relatively stable over time.
Conclusions
TB diagnosis delay may influence recovery beyond treatment success, highlighting the importance of early case detection and integrated post-treatment support in high-burden settings.
Keywords: Tuberculosis, Patient Outcome Assessment
WHAT IS ALREADY KNOWN ON THIS TOPIC.
WHAT THIS STUDY ADDS
This prospective cohort study provides longitudinal evidence from Cambodia on the association between TB diagnosis delay and trajectories of respiratory health, quality of life and depressive symptoms.
Individuals with longer diagnosis delays experienced worse respiratory health and lower psychological and environmental quality of life, both before treatment and during follow-up, despite overall recovery in both groups.
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY
Strengthening early case detection and integrating post-TB assessment and supportive care into routine TB services may help reduce persistent health disparities following treatment.
Background
Tuberculosis (TB) continues to be a major infectious disease, affecting 10.7 million people and claimed the life of 1.23 million people in 2024.1 Pulmonary TB is the most common form and the primary contributor to transmission and long-term health consequences.2 TB is considered a curable disease, with a treatment success rate of 88% globally in 2023.3 Nevertheless, increasing evidence shows that many TB survivors experience persistent respiratory impairment, functional limitations and reduced well-being even after successful treatment.4–6 These findings challenge the traditional focus of TB programmes on short-term treatment outcomes and highlight the need to consider longer-term health consequences among TB survivors.
Recognition of these issues led to growing international efforts to define and address post-tuberculosis (post-TB) morbidity. The First International Post-Tuberculosis Symposium, held in 2019 in Stellenbosch, South Africa, introduced a broad consensus definition of post-TB lung health as “evidence of chronic respiratory abnormality, with or without symptoms, attributable at least in part to previous pulmonary TB”.7 More recently, a research case definition for post-TB lung disease was proposed, requiring a person to have a history of pulmonary or pleural TB without active TB at the assessment time, and present with evidence of pulmonary disease in at least two of three domains: lung function, respiratory symptoms and chest imaging, and attribution at least partly to previous TB.8 Subsequent research has demonstrated that the burden of post-TB health is substantial. A hypothetical global cohort analysis estimated that approximately 58 million out of 122 million of the disability-adjusted life years (DALYs) attributable to TB were due to post-TB health rather than the acute disease episode itself.9 Similarly, a systematic review reported that TB survivors experience almost threefold higher mortality compared with matched general population controls, highlighting the long-term risks faced by this population.10 Despite this evidence, post-TB care remains unintegrated into routine TB programmes.
Post-TB health is possibly aggravated by TB diagnosis and treatment delay. A recent microsimulation modelling study estimated that TB treatment delay would result in a greater reduction in life expectancy (7.7 years vs 5.1 years) and greater lifetime health losses (17.1 DALYs vs 11.4 DALYs), which may be driven mainly by post-TB sequelae.11 Similarly, a narrative review identified delayed or incomplete TB treatment as a key risk factor for persistent post-TB respiratory morbidity and impaired quality of life.12 However, despite emerging longitudinal studies suggesting that respiratory impairment can persist beyond 12 months after treatment completion, investigations linking diagnosis delay to subsequent long-term respiratory health and broader well-being remain limited.
Estimating the impact of delayed TB diagnosis on long-term respiratory health and well-being is critical, as diagnosis delay is a modifiable component of the TB care pathway that may affect outcomes beyond treatment completion. TB control programmes have traditionally prioritised treatment success and microbiological cure, while post-TB health, such as persistent respiratory impairment and reduced quality of life, has not received due attention. Understanding whether TB diagnosis delay contributes to these longer-term outcomes can inform strategies to strengthen early case detection and support the integration of post-TB care into routine TB services. Therefore, this study examines the associations between TB diagnosis delay and respiratory health, quality of life and depressive symptoms from the time of diagnosis through 2 years of follow-up in a cohort of TB survivors in Cambodia, one of the high-TB-burden countries.
Methods
Study design and settings
This prospective cohort study was conducted within the Community Mobilisation Initiatives to End Tuberculosis (COMMIT) project, managed by the Khmer HIV/AIDS NGO Alliance (KHANA) in Cambodia. The cohort sites included ten operational districts (ODs) in four provinces, namely: Stung Trang and Kang Meas in Kampong Cham Province; Ou Reang Ov and Suong in Tbong Khmum Province; Muk Kampoul, Lvea Em, Sa Ang and Leuk Dek in Kandal Province and Por Sen Chey and Sen Sok in the capital of Phnom Penh (figure 1).
Figure 1. COMMIT Project operational sites in Cambodia. COMMIT, Community Mobilisation Initiatives to End Tuberculosis; ODs, operational districts.
Participants
Adults aged ≥18 years with newly diagnosed TB were enrolled in the cohort. The baseline survey was conducted within 1 month of participants receiving their TB diagnosis. They were recruited through various interventions in the COMMIT project, including active case finding (ACF) using a seed-and-recruit model, one-off roving ACF, hospital linkages and bidirectional screening for TB and diabetes. In the present study, which aims to examine the health outcomes of TB diagnosis delay, we excluded those who reported no TB symptoms before diagnosis (see the definition in the Variables and measurement section) because they had no diagnosis delay.
Data collection
Structured questionnaires were administered in person (or by phone, depending on the situation during the COVID-19 pandemic) by trained data collectors. Data collectors received training from researchers at the KHANA Centre for Population Health Research (KHANA-CPHR). Data were collected electronically using tablets via REDCap (Research Electronic Data Capture), hosted on a secure server at KHANA-CPHR. Participants were identified from a list of individuals diagnosed with TB through the COMMIT project. Data collectors coordinated with COMMIT field staff and contacted potential participants by phone or through field staff to invite them to participate. Prior to enrolment, data collectors explained the study and obtained written informed consent. Participants could choose their preferred interview location (such as health centres, referral hospitals or their homes), and interviews were conducted in a private setting to ensure confidentiality. All participants received a US$ five reimbursement for their time and participation.
The baseline data were collected in 2022, with two follow-up surveys conducted in 2023 and 2024 (1 year intervals). Participants who did not complete one follow-up survey remained eligible for contact and participation at the subsequent follow-up survey. Identifiable information was accessible only to authorised research staff for follow-up and was not used in the analyses.
Variables and measurements
An exposure variable is TB diagnosis delay, defined as the duration between the onset of TB symptoms and the date of TB diagnosis.13 The onset of TB symptoms, including cough, haemoptysis, chest pain, dyspnoea, fever, chills, weight loss and night sweats, was based on participants’ recall. The date of TB diagnosis was recorded during ACF interventions in the COMMIT project. Because no universally accepted clinical threshold for prolonged TB diagnosis delay exists, participants were categorised using the cohort median of 64 days. Participants with a delay of 64 days or longer were classified as having a long diagnosis delay, whereas those with a delay shorter than 64 days were classified as having a short diagnosis delay. This median split was used to compare participants with relatively longer and shorter delays and was not intended to represent a clinically validated threshold.14 15
The outcome was respiratory health status, assessed using the St George’s Respiratory Questionnaire for Chronic Obstructive Pulmonary Disease (SGRQ-C).16 The SGRQ-C is a validated patient-reported measure and comprises three domains: symptoms, activity and impacts. The Symptoms domain captures the frequency and severity of respiratory symptoms, including cough, wheeze and breathlessness. The Activity domain reflects limitations in physical activity due to breathlessness. The Impacts domain assesses the social and psychological consequences of respiratory disease, including effects on daily functioning, employment and emotional well-being. Domain-specific and total scores were calculated using the developer’s weighted scoring algorithm and transformed to a scale ranging from 0 to 100, with higher scores indicating worse respiratory health.
Additional outcomes include quality of life and depressive symptoms. Quality of life was measured using the WHO Quality of Life-BREF (WHOQOL-BREF), a 26-item instrument covering four domains: physical health, psychological health, social relationships and environment.17 18 The Physical Health domain assesses mobility, energy, pain and capacity for daily activities. The Psychological Health domain assesses emotional well-being, self-esteem and cognitive functioning. The Social Relationships domain assesses personal relationships, social support and social interactions. The Environment domain assesses broader contextual factors including financial resources, access to health services, safety and living conditions. Items were rated on a five-point Likert scale and domain scores were calculated and transformed in accordance with WHO guidelines with higher scores indicating better quality of life. Depressive symptoms were assessed using the 10-item Centre for Epidemiologic Studies Depression Scale (CES-D-10)19 which measures depressive symptoms experienced during the past week. Responses were scored on a four-point Likert scale yielding a total score ranging from 0 to 30; a score of 10 or higher indicated the presence of depressive symptoms.
Other variables included sociodemographic characteristics (sex, age, education, monthly household income and economic status) and clinical and behavioural factors (TB type and treatment status, cigarette smoking and alcohol use). Economic status was defined by comparing participants’ per-person per-day income with the national poverty line set by the Royal Government of Cambodia which is US$2.69 (KHR 10 951) per person per day. We categorised them as “Poor”, “Near-poor” and “Non-poor”, where “Poor” refers to those with income below the poverty line, “Near-poor” to those with income between 100% and 125% of the poverty line and “Non-poor” to those with income above 125% of the poverty line.20 Income per person per day was calculated by dividing household monthly income by the number of household members and averaging over 30 days per month.
The exposure variable (TB diagnosis delay) was assessed at baseline. All outcome variables were assessed at all waves, except types of TB, treatment status, smoking and alcohol use, which were assessed only at baseline.
Statistical analysis
First, we described the participants’ characteristics by the duration of delayed TB diagnosis: short- (<64 days) and long-delay (≥64 days) groups. Then, we examined the associations between TB diagnosis delay and outcome variables, using linear mixed-effects models (LMMs), with random intercepts specified at the participant level to account for within-individual correlation arising from repeated measurements. Time was modelled as a categorical variable with baseline as the reference category. Models included an interaction between diagnosis-delay group and follow-up time to examine whether the difference between long- and short-delay groups changed at the 1 year and 2 year follow-up surveys. The models were adjusted for potential confounders, including age, sex, baseline economic status, TB type, cigarette smoking and alcohol use. Results are reported as adjusted regression coefficients with 95% CIs. Full fixed-effects estimates, including covariates and interaction terms, are presented in online supplemental tables S1–S9).
Participants who were reported as deceased during follow-up and those who were lost to both follow-up surveys were excluded from the analyses (n=156, 22%). However, participants who completed only one follow-up survey were retained (n=77, 13.8%). LMMs used all available repeated-outcome observations under the assumption that missing outcome data were missing at random.
At baseline, 22 participants (3.9%) had missing SGRQ-C scores. At the first and second follow-up surveys, 46 (8.3%) and 31 (5.6%) participants, respectively, had missing outcome data. There was no missing data for the diagnosis-delay group or covariates included in the models. As a sensitivity analysis, complete-case LMMs were fitted among participants who completed both follow-up surveys (n=479). All statistical analyses were conducted using R (V.4.3.1, The R Foundation for Statistical Computing).
Patient and public involvement
The COMMIT cohort study was conducted in collaboration with multiple stakeholders involved in TB control in Cambodia. The study was implemented in partnership with the National TB Programme, local health authorities and organisations supporting ACF activities in the COMMIT project. Community health workers and TB healthcare providers at referral hospitals and health centres in COMMIT selected ODs assisted in identifying and facilitating contact with individuals diagnosed through ACF activities. Local authorities and community members also played an important role in fostering community engagement and participation in ACF events, thereby facilitating recruitment into the cohort. Patients and community members were not involved in the study design or analysis but participated as respondents in the cohort surveys.
Results
Characteristics of participants
Of 712 participants in the baseline survey, 53% were males, with a median age of 59 years old (range: 18–91), 75% had received no or primary education only and 79% were classified as having poor economic status, with a median monthly household income of US$215 (range: 0–1900). Regarding TB, 41% were diagnosed with bacteriologically confirmed TB, while 55% were diagnosed with clinically diagnosed TB. Most of the study participants were successfully treated for TB (n=685, 96.2%). In the first and second follow-up surveys, 526 and 525 participants (74%) were retained, respectively.
Table 1 compares participant characteristics at each time point between the long- and short-delay groups. The median duration between the onset of TB symptoms and the date of TB diagnosis was 119 days (IQR: 87 to 180) in the long-delay group and 31 days (14–46) in the short-delay group. At baseline, the characteristics were generally similar between the long- and short-delay groups, except for sex, education, economic status and comorbidity. The proportions of men, individuals with education beyond primary school and non-poor individuals were slightly lower in the long-delay group than in the short-delay group. Such differences largely persisted at follow-ups.
Table 1. Demographic, socio-economic and clinical characteristics of participants at each survey time-point in the long- and short-delay groups.
| Variables | Baseline (n=712) | First follow-up (n=526) | Second follow-up (n=525) | |||
|---|---|---|---|---|---|---|
| Long delay | Short delay | Long delay | Short delay | Long delay | Short delay | |
| (n=359) | (n=353) | (n=267) | (n=259) | (n=271) | (n=254) | |
| Sex (Male) | 172 (47.9%) | 204 (57.8%) | 126 (47.2%) | 145 (56.0%) | 127 (46.9%) | 139 (54.7%) |
| Age (Median, IQR) | 60 (18 to 91) | 58 (19 to 87) | 62 (19 to 87) | 60 (23 to 87) | 62 (20 to 88) | 61 (22 to 88) |
| Education | ||||||
| Primary and below | 281 (78.3%) | 254 (72%) | 212 (79.4%) | 197 (76.1%) | 210 (77.5%) | 192 (75.6%) |
| Above primary | 78 (21.7%) | 99 (28%) | 55 (20.6%) | 62 (23.9%) | 61 (22.5%) | 62 (24.4%) |
| Monthly household income (US$) (Median, IQR) | 200 (10 to 1100) | 225 (0 to 1900) | 190 (0 to 1050) | 160 (0 to 1050) | 200 (26.7 to 1000) | 200 (25 to 1000) |
| Economic status* | ||||||
| Poor | 293 (81.6%) | 271 (76.8%) | 228 (85.4%) | 211 (81.5%) | 228 (84.1%) | 199 (78.3%) |
| Near poor | 31 (8.6%) | 26 (7.4%) | 20 (7.5%) | 23 (8.9%) | 20 (7.4%) | 25 (9.8%) |
| Non-poor | 35 (9.7%) | 56 (15.9%) | 19 (7.1%) | 25 (9.7%) | 23 (8.5%) | 30 (11.8%) |
| Type of TB† | ||||||
| Bacteriologically confirmed TB | 147 (40.9%) | 148 (41.9%) | 100 (37.5%) | 96 (37.1%) | 99 (36.5%) | 90 (35.4%) |
| Clinically diagnosed TB | 196 (54.6%) | 193 (54.7%) | 157 (58.8%) | 155 (59.8%) | 162 (59.8%) | 153 (60.2%) |
| Extra-pulmonary TB | 16 (4.5%) | 12 (3.4%) | 10 (3.7%) | 8 (3.1%) | 10 (3.7%) | 11 (4.3%) |
| Treatment status | ||||||
| Completed | 205 (57.1%) | 199 (56.4%) | ||||
| Cured | 142 (39.6%) | 146 (41.4%) | ||||
| Died | 4 (1.1%) | 4 (1.1%) | ||||
| Lost to follow-up | 8 (2.2%) | 4 (1.1%) | ||||
| Cigarette smoking | ||||||
| Current smokers | 41 (11.4%) | 44 (12.5%) | ||||
| Former smokers | 99 (27.6%) | 122 (34.6%) | ||||
| Non-smokers | 219 (61%) | 187 (53%) | ||||
| Alcohol use | ||||||
| 4 times or more a week | 32 (8.9%) | 32 (9.1%) | ||||
| 2–3 times per week | 19 (5.3%) | 17 (4.8%) | ||||
| 2–4 times per week | 22 (6.1%) | 25 (7.1%) | ||||
| Once a month or less | 8 (2.2%) | 10 (2.8%) | ||||
| None | 278 (77.4%) | 269 (76.2%) | ||||
Economic status was defined using the national poverty line set by the Royal Government of Cambodia, an income of US$ 2.69 (KHR 10 951) per person per day. The poor are those living below the poverty line. Near poor are those living from 100% to 125% of the poverty line. Non-poor are those living above 125% of the poverty line.
Type of TB was recorded at baseline. Values shown in the follow-up columns represent the baseline TB type of participants who completed the respective follow-up survey and do not indicate reassessment or change in TB type.
IQR, Interquartile Range; n, Frequency; TB, tuberculosis; USD, United States Dollar.
More than two-thirds of participants (67.3%) completed both follow-up surveys. At the first follow-up, 73.9% completed the survey; 186 did not participate, including deaths (3.2%), refusals (6.7%) and losses to follow-up (16.2%). At the second follow-up, 73.7% completed the survey. Non-participation at the second follow-up included deaths occurring during both follow-ups (5.9%), refusals (6.2%) and losses to follow-up (14.2%). Because some participants missed one follow-up survey but returned at the next wave, the number of participants completing each follow-up survey does not follow a strictly sequential pattern. Participants who did not complete both follow-up surveys were younger on average (48.4 years vs 59.6 years) and had a lower proportion of females (37.7% vs 50.1%) than those who completed both surveys, whereas the proportion of participants with a long diagnosis delay was similar in both groups (53.2% vs 50.9%).
TB symptoms, respiratory health, quality of life and depressive symptoms
Table 2 compares TB symptoms, respiratory health, quality of life and depressive symptoms at each time point between the long- and short-delay groups. At baseline, the long-delay group exhibited more TB symptoms, had poorer respiratory health scores and slightly lower quality of life scores than the short-delay group. However, differences in the depressive symptom scores and the prevalence of depressive symptoms were minimal between the two groups. The participants’ conditions were improving over time in both groups after treatment completion. However, the long-delay group consistently reported having worse outcomes than the short-delay group. Notably, TB symptoms persisted in both groups even at the second follow-up: shortness of breath in 14.2% and 7.6%, and wheezing attacks in 15% and 9.9% for the long- and short-delay groups, respectively.
Table 2. TB symptoms, respiratory health, quality of life and depressive symptoms in the long- and short-delay groups at the baseline and follow-up surveys.
| Variables | Baseline | First follow-up | Second follow-up | |||
|---|---|---|---|---|---|---|
| Long delay | Short delay | Long delay | Short delay | Long delay | Short delay | |
| (n=359) | (n=353) | (n=267) | (n=259) | (n=271) | (n=254) | |
| TB symptoms | ||||||
| Cough | 274 (76.3%) | 256 (72.5%) | 92 (25.6%) | 96 (27.2%) | 88 (24.5%) | 91 (25.8%) |
| Cough with phlegm (sputum) | 199 (55.4%) | 174 (49.3%) | 76 (21.2%) | 64 (18.1%) | 62 (17.3%) | 68 (19.3%) |
| Shortness of breath | 215 (59.9%) | 173 (49.0%) | 56 (15.6%) | 44 (12.5%) | 51 (14.2%) | 27 (7.6%) |
| Attack of wheezing | 135 (37.6%) | 127 (36.0%) | 57 (15.9%) | 49 (13.9%) | 54 (15.0%) | 35 (9.9%) |
| Respiratory health (Mean, SD)† | ||||||
| Symptoms score | 42.6 (22.8) | 36.5 (23.4) | 22.5 (20.6) | 19.5 (19.3) | 12.9 (20.1) | 10.5 (17.8) |
| Activity score | 48.7 (31.2) | 41.5 (32.0) | 20.5 (25.3) | 17.9 (24.0) | 17.2 (23.0) | 11.1 (17.7) |
| Impact score | 29.1 (26.7) | 25.9 (24.6) | 11.8 (18.9) | 10.6 (18.0) | 9.53 (15.6) | 6.48 (12.4) |
| Total score | 37.5 (24.1) | 32.5 (23.7) | 16.4 (19.5) | 14.4 (18.5) | 12.5 (17.4) | 8.6 (13.9) |
| Quality of life (Mean, SD)‡ | ||||||
| Physical health score | 60.5 (14.3) | 62.6 (13.7) | 69.1 (14.1) | 71.3 (13.7) | 69.4 (13.2) | 71.1 (12.0) |
| Psychological health score | 59.3 (15.2) | 63.2 (14.1) | 66.5 (13.8) | 69.2 (13.0) | 68.1 (12.0) | 70.6 (10.8) |
| Social relationships score | 60.6 (10.9) | 61.7 (11.4) | 62.7 (10.0) | 62.3 (9.85) | 63.4 (7.8) | 64.6 (8.3) |
| Environment score | 56.5 (9.2) | 59 (9.3) | 57.5 (8.6) | 58.5 (8.8) | 55.9 (7.5) | 56.9 (7.7) |
| Depressive symptoms§ | ||||||
| CES-D-10 score (Mean, SD) | 12 (6.1) | 11.6 (5.4) | 10.6 (5.0) | 9.75 (4.3) | 10 (4.4) | 9.39 (3.6) |
| With depressive symptoms | 197 (54.9%) | 189 (53.5%) | 124 (34.5%) | 96 (27.2%) | 99 (27.6%) | 89 (25.2%) |
TB symptoms derived from the SGRQ-C questionnaire: I (cough/cough with phlegm (sputum)/have shortness of breath/have attacks of wheezing): most days of the week/several days a week/only with chest infection/not at all.
Respiratory health is measured by the SGRQ-C questionnaire. We presented the three domains’ scores of SGRQ-C and its total score (range: 0–100, lower is better).
Quality of life is measured by the WHOQOL-BREF questionnaire. We presented the four domains’ scores of WHOQOL-BREF (range: 0–100, higher is better).
Depressive symptoms are measured by the CESD-10 questionnaire. We presented the number of those who were categorised as depressed (CES-D-10 score ≥10) and their mean scores (range: 0–30, lower is better).
CES-D-10, Center of Epidemiologic Studies Depression Scale,10-item version; n, Frequency; SD, Standard deviation.
Associations of TB diagnosis delay with respiratory health, quality of life and depressive symptoms
In adjusted LMMs, long TB diagnosis delay was associated with worse respiratory health over the 2 year follow-up period (table 3; online supplemental tables S1–S4); online supplemental figure 1). Compared with the short-delay group, the long-delay group had higher SGRQ-C total, symptom and activity scores. There was no evidence that the adjusted differences between long- and short-delay groups changed at the 1-year or 2-year follow-up visits, as none of the diagnosis-delay-by-time interaction terms were statistically significant. For quality of life, long TB diagnosis delay was significantly associated with lower psychological health and environment scores but not with physical health and social relationships scores (online supplemental tables S5–S8; online supplemental figure 2). Long TB diagnosis delay was not associated with a depressive symptom score (online supplemental table S9; online supplemental figure 3). No significant interactions between TB diagnosis delay and time were observed for both quality of life and depressive symptom scores. Complete-case analyses yielded results similar to the primary analyses in both direction and magnitude. However, estimates were less precise due to reduced sample size (online supplemental tables S10–S18).
Table 3. Linear mixed-effects models examining the association of TB diagnosis delay with respiratory health, quality of life and depressive symptoms over 2 year of follow-ups.
| Outcome | Adjusted β (95% CI)* for diagnosis delay (Ref=Short delay) | |
|---|---|---|
| Main analysis models | Sensitivity analysis models† | |
| Respiratory health‡ | ||
| Total score | 4.37 (1.10 to 7.64) | 4.02 (0.51 to 7.52) |
| Symptoms score | 5.83 (2.34 to 9.31) | 5.70 (1.95 to 9.44) |
| Activity score | 6.95 (2.67 to 11.23) | 7.01 (2.46 to 11.56) |
| Impact score | 2.38 (−0.9 to 5.66) | 1.70 (−1.82 to 5.23) |
| Quality of life§ | ||
| Physical health score | −1.70 (−3.84 to 0.43) | −1.72 (−4.03 to 0.60) |
| Psychological health score | −3.17 (−5.31 to −1.03) | −3.26 (−5.56 to −0.96) |
| Social relationships score | −0.94 (−2.54 to 0.66) | −0.6 (−2.3 to 1.1) |
| Environment score | −1.74 (−3.07 to −0.4) | −1.75 (−3.19 to −0.3) |
| Depressive symptoms¶ | −0.12 (−0.91 to 0.66) | −0.24 (−1.10 to 0.62) |
Models were adjusted for age, sex, baseline economic status, TB type, cigarette smoking and alcohol use. Random intercepts were specified at the participant level, and time (baseline, 1-year and 2-year follow-up) and its interaction with diagnosis delay were included in the models.
Complete-case analysis was performed for the sensitivity analysis (n = 479).
Respiratory health is measured by the SGRQ-C questionnaire. We presented the three domains’ scores of SGRQ-C and its total score (range: 0–100, lower is better).
Quality of life is measured by the WHOQOL-BREF questionnaire. We presented the four domains’ scores of WHOQOL-BREF (range: 0–100, higher is better).
Depressive symptoms are measured by the CESD-10 questionnaire. We presented the score as a continuous variable (range: 0–30, lower is better).
CI, confidence interval; Ref, reference; β, the regression coefficient from linear mixed-effects model.
Discussion
In this prospective cohort study of TB survivors in Cambodia, we observed that long TB diagnosis delay was associated with worse patient-reported respiratory health and some domains of quality of life from baseline through 2 years of follow-up. These findings suggest that delayed TB diagnosis may have consequences in respiratory symptom burden, activity limitation and well-being that extend beyond treatment completion.
TB diagnosis delay and respiratory health
In this cohort, longer TB diagnosis delay was associated with worse patient-reported respiratory health, including higher SGRQ-C total, symptoms and activity scores than in the short-delay group. Respiratory health improved substantially in both delay groups during follow-up. However, the differences in patient-reported respiratory health between participants with long and short diagnosis delay did not change at the 1-year or 2-year follow-up visits. These findings suggest that the respiratory health disadvantage observed among participants with longer diagnosis delays around the time of TB diagnosis was not fully eliminated over the subsequent 2 years of follow-up.
TB diagnosis delay reflects prolonged, untreated pulmonary disease, which may contribute to more severe respiratory symptoms such as haemoptysis, lung damage and airway damage. Previous work has linked delayed treatment with chronic airflow obstruction, and there are studies that suggest that prompt TB diagnosis and treatment may reduce long-term health losses associated with progressive lung damage and post-TB sequelae.11 21 However, because our study did not include spirometry or chest imaging, we could not determine whether structural or physiological lung abnormalities explained the observed differences in patient-reported respiratory health between the two groups. Our findings should not be interpreted as evidence of post-TB lung disease as a clinical diagnosis.
Our findings are consistent with previous studies showing that respiratory morbidity can persist after successful TB treatment. Reviews and clinical studies have described chronic cough, dyspnoea, activity limitation and impaired respiratory quality of life among TB survivors.12 22–24 Importantly, emerging longitudinal evidence suggests that recovery may continue beyond treatment completion but is often incomplete. In a prospective adult cohort in Malawi, respiratory symptoms and lung function improved mainly during the first year after pulmonary TB treatment, although respiratory symptoms and abnormal spirometry remained common at 3 years.25 Similarly, a 24-month prospective study of adolescent TB survivors in Peru found that selected lung function measures improved over time, whereas respiratory disability, measured using the SGRQ, remained worse than healthy adolescents with no TB history.26 While these studies differ from ours in population group, settings, timing of assessment and outcome measures, they presented a common pattern of partial respiratory recovery after TB treatment with persistent residual burden for some individuals. Our study extends this evidence by examining diagnosis delay as an underlying risk factor and by following adults from shortly after TB diagnosis through 2 years of follow-up. Together, the findings suggest that earlier TB diagnosis may be important not only for reducing acute morbidity and transmission but also for reducing longer-term disparities in patient-reported respiratory health.
TB diagnosis delay and quality of life and depressive symptoms
Participants in the long TB diagnosis delay group had lower quality of life, particularly in the psychological and environmental domains, than those in the short-delay group. This is plausible because previous studies in post-TB populations showed that persistent respiratory symptoms and functional impairment are closely linked to poorer mental well-being and reduced quality of life in the psychological domain.27 28 Individuals who experienced prolonged TB symptoms before diagnosis may also have faced greater uncertainty, distress and disruption to daily functioning,15 which could have longer-term implications for psychological health.
The environment domain of quality of life encompasses financial resources, living conditions and access to services. It may be particularly sensitive to prolonged untreated illness because TB survivors often report ongoing challenges in returning to work and restoring daily activities after treatment.29 Therefore, it is likely that participants with longer periods of TB symptoms before diagnosis faced greater and more sustained economic and daily life constraints than those diagnosed earlier. Ultimately, our findings suggest that TB diagnosis delay may contribute to sustained disparities in well-being between those diagnosed earlier and later beyond respiratory health alone.
However, among our participants, TB diagnosis delay was not associated with the physical-health or social-relationship domains of quality of life or with depressive symptoms score. These null findings may indicate that the association between diagnosis delay and well-being varies across domains. However, they should not be interpreted as evidence that delayed diagnosis has no physical and psychosocial consequences. Physical health may have improved similarly in both groups after treatment, while social relationships and depressive symptoms may be influenced by factors not measured in this study, including household support, stigma and access to care. Nevertheless, depressive symptoms remained high at the 2-year follow-up in both diagnosis-delay groups, affecting 27.6% of participants with long delay and 25.2% of those with short delay. This finding highlights an ongoing need for mental-health assessment and support among TB survivors, irrespective of diagnosis-delay duration.
Policy implication
Our findings support extending TB care beyond treatment completion. In settings such as Cambodia and many other lower-middle-income countries, where routine health check-ups are uncommon, national TB programmes could consider introducing longer-term follow-up for TB survivors that assesses persistent respiratory symptoms, activity limitation, quality of life and mental health needs, with referral for further respiratory evaluation, rehabilitation or psychosocial support when indicated.30 This is particularly important for people with prolonged diagnosis delay, who may enter treatment with greater respiratory health disadvantage. At the same time, depressive symptoms remained common at 2 years in both delay groups, indicating that mental health assessment and support should not be limited to those with long diagnosis delay. Integrating these components into routine TB services could help identify persistent needs after microbiological cure and promote a more person-centred approach to TB care.
We also found that the proportion of women and participants with lower education and economic status was higher in the long-delay group. Strengthening early case detection should therefore include approaches accessible to people who may face financial, informational or gender-related barriers to care. Community-based case finding, prompt symptom assessment, clear referral pathways and efforts to reduce the indirect costs of accessing TB services may help shorten the interval between symptom onset and diagnosis.31 Although our observational findings do not establish causality, they support the potential value of earlier diagnosis in reducing disparities in subsequent respiratory health and well-being.
Study limitations
Several limitations should be considered when interpreting these findings. First, key exposure and outcome variables, including TB symptom onset, respiratory health, quality of life and depressive symptoms, were based on self-reported data and may be subject to recall or social desirability bias. In particular, the timing of TB symptom onset may be imperfectly recalled, especially among the study participants with prolonged illness. To minimise recall bias, data collectors asked them to recall the onset of specific TB symptoms using disaggregated symptom questions and employed temporal reference points, such as notable personal or public events, to aid recall. This approach has been shown to improve recall accuracy through prompting and clarification.32
Second, the study participants were recruited through TB ACF programmes, which facilitate earlier TB diagnosis and treatment and may provide psychosocial support through ACF-related activities. Therefore, TB diagnosis delay could have been shorter among our study participants, and the association between TB diagnosis delay and psychosocial outcomes might have been attenuated.
Third, 22% of participants were deceased or lost to follow-up and were therefore excluded from the analyses. In our sensitivity analyses following a complete-case approach, the association between TB diagnosis delay and outcome variables remains unchanged. Moreover, the characteristics of the long- and short-delay groups at baseline did not differ substantially between the first and second follow-ups, suggesting that deaths and losses to follow-up occurred at random.
Fourth, information on key comorbid conditions was limited. HIV was recorded for only four participants and could not be reliably included in adjusted models. We also lacked standardised longitudinal data on chronic respiratory comorbidities, such as asthma and chronic obstructive pulmonary disease, as well as other non-communicable conditions that may influence respiratory health over time. Because such conditions may affect not only respiratory symptoms but also the timing of diagnosis, and because they were not fully measured, residual confounding cannot be excluded.
Fifth, although we used validated instruments, respiratory health was self-reported, without objective assessment of structural lung damage using clinical tests such as spirometry and imaging, due to limited resources. Consequently, we could not assess lung function impairment, identify structural lung abnormalities, determine whether worse SGRQ-C scores reflected TB-related physiological impairment or structural lung damage, or classify participants according to the recently proposed research definition of post-TB lung disease.8 Therefore, our findings should be interpreted as associations between TB diagnosis delay and patient-reported respiratory symptom burden and activity limitation rather than evidence of clinically defined post-TB lung disease or structural pulmonary sequelae. In addition, respiratory symptoms captured by the SGRQ-C are nonspecific and may not be exclusively attributable to prior TB, particularly at follow-up after treatment completion. Symptom reporting is inherently subjective and may vary with individual perception and context, including awareness of diagnosis.33 Nevertheless, patient-reported outcomes remain important because they capture symptoms, functional limitation, and the day-to-day consequences of illness experienced by TB survivors. In Cambodia, where post-TB health and well-being have been little described, these findings provide an important initial characterisation of longitudinal respiratory health and well-being among TB survivors.
Finally, a causal relationship between exposure and outcome variables remains uncertain due to the observational nature of this study. However, the longitudinal study design helped ensure a temporal relationship between them.
Conclusion
Long TB diagnosis delay was associated with worse patient-reported respiratory health and lower psychological and environmental domains of quality of life from baseline through 2 years of follow-up among TB survivors in Cambodia. These findings suggest that prolonged time to TB diagnosis may be associated with persistent disparities in respiratory symptom burden, activity limitation and selected dimensions of well-being beyond TB treatment. Strengthening early case detection and timely TB diagnosis, alongside assessment of persistent respiratory and psychosocial needs after treatment, may help support more person-centred TB care. Future post-TB-related studies should incorporate spirometry and chest imaging with patient-reported outcomes to clarify the clinical correlates and mechanisms of these associations.
Supplementary material
Acknowledgements
We are grateful for the support of people and communities affected by TB and acknowledge the difficulties they have experienced as a consequence of the disease. We would also like to thank the healthcare workers, community volunteers, field staff and data collectors for their contributions. We extend special thanks to the National Centre for Tuberculosis and Leprosy Control, the United States Agency for International Development and the Cambodian Committee for TB Research.
Footnotes
Funding: COMMIT Cohort study is funded by the United States Agency for International Development (USAID).
Provenance and peer review: Not commissioned; externally peer reviewed.
Patient consent for publication: Not applicable.
Ethics approval: This study involves human participants and was approved by this cohort study was approved by the National Ethics Committee for Health Research in Cambodia (Reference number: 319/NECHR in 2022, 051/NECHR in 2023 and 070/NECHR in 2024) and the Research Ethics Committee of the Institute of Medicine at the University of Tsukuba (No. 2167) in September 2025. Participants gave informed consent to participate in the study before taking part.
Data availability free text: Data may be available upon request. Requests for anonymised data or collaboration inquiries may be made to the corresponding author.
Map disclaimer: The depiction of boundaries on this map does not imply the expression of any opinion whatsoever on the part of BMJ (or any member of its group) concerning the legal status of any country, territory, jurisdiction or area or of its authorities. This map is provided without any warranty of any kind, either express or implied.
Patient and public involvement: Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.
Data availability statement
Data are available upon reasonable request.
References
- 1.World Health Organization Tuberculosis: key facts. [6-Jun-2026]. https://www.who.int/news-room/fact-sheets/detail/tuberculosis Available. Accessed.
- 2.Centers for Disease Control and Prevention About tuberculosis. [21-Dec-2025]. https://www.cdc.gov/tb/about/?CDC_AAref_Val=https://www.cdc.gov/tb/topic/basics/default.htm Available. Accessed.
- 3.World Bank Group Tuberculosis treatment success rate (% of new cases) [21-Feb-2025]. https://data.worldbank.org/indicator/SH.TBS.CURE.ZS Available. Accessed.
- 4.World Health Organization Policy brief on tuberculosis- associated disability. 2023. https://iris.who.int/bitstream/handle/10665/373679/9789240077799-eng.pdf Available.
- 5.Gai X, Cao W, Rao Y, et al. Risk factors and biomarkers for post-tuberculosis lung damage in a Chinese cohort of male smokers and non-smokers: protocol for a prospective observational study. BMJ Open. 2023;13:e065990. doi: 10.1136/bmjopen-2022-065990. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Pontali E, Centis R. TB survivors: why curing TB is sometimes not enough. IJTLD Open . 2024;1:531–2. doi: 10.5588/ijtldopen.24.0530. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Allwood BW, van der Zalm MM, Amaral AFS, et al. Post-tuberculosis lung health: perspectives from the First International Symposium. Int J Tuberc Lung Dis. 2020;24:820–8. doi: 10.5588/ijtld.20.0067. [DOI] [PubMed] [Google Scholar]
- 8.Bisson GP, Allwood B, Byrne A, et al. Post-tuberculosis lung disease: a case definition for use in research studies. Lancet Infect Dis. 2026;26:e315–25. doi: 10.1016/S1473-3099(25)00548-1. [DOI] [PubMed] [Google Scholar]
- 9.Menzies NA, Quaife M, Allwood BW, et al. Lifetime burden of disease due to incident tuberculosis: a global reappraisal including post-tuberculosis sequelae. Lancet Glob Health. 2021;9:e1679–87. doi: 10.1016/S2214-109X(21)00367-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Romanowski K, Baumann B, Basham CA, et al. Long-term all-cause mortality in people treated for tuberculosis: a systematic review and meta-analysis. Lancet Infect Dis. 2019;19:1129–37. doi: 10.1016/S1473-3099(19)30309-3. [DOI] [PubMed] [Google Scholar]
- 11.Can MH, Sweeney S, Allwood BW, et al. Implications of progressive lung damage and post-tuberculosis sequelae for the health benefits of prompt tuberculosis treatment in high HIV prevalence settings: a mathematical modelling analysis. Lancet Glob Health. 2025;13:e1240–9. doi: 10.1016/S2214-109X(25)00114-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Wang J, Yuan B, Fang Y, et al. Post-tuberculosis morbidities and their associated mortality: moving from challenges to solutions. Eur Respir Rev. 2025;34:250148. doi: 10.1183/16000617.0148-2025. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Paramasivam S, Thomas B, Chandran P, et al. Diagnostic delay and associated factors among patients with pulmonary tuberculosis in Kerala. J Family Med Prim Care. 2017;6:643–8. doi: 10.4103/2249-4863.222052. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Sreeramareddy CT, Panduru KV, Menten J, et al. Time delays in diagnosis of pulmonary tuberculosis: a systematic review of literature. BMC Infect Dis. 2009;9:91. doi: 10.1186/1471-2334-9-91. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Teo AKJ, Singh SR, Prem K, et al. Duration and determinants of delayed tuberculosis diagnosis and treatment in high-burden countries: a mixed-methods systematic review and meta-analysis. Respir Res. 2021;22:251. doi: 10.1186/s12931-021-01841-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Jones PW, Forde Y. St George’s, University of London; [25-May-2025]. ST george’S respiratory questionnaire for copd patients (sgrq-c): manual.https://www.sgul.ac.uk/research/research-operations/research-administration/st-georges-respiratory-questionnaire/docs/SGRQ-C-Manual-January-2023.pdf Available. Accessed. [Google Scholar]
- 17.The Whoqol Group Development of the World Health Organization WHOQOL-BREF Quality of Life Assessment. Psychol Med. 1998;28:551–8. doi: 10.1017/S0033291798006667. [DOI] [PubMed] [Google Scholar]
- 18.Chung W-S, Lan Y-L, Yang M-C. Psychometric testing of the short version of the world health organization quality of life (WHOQOL-BREF) questionnaire among pulmonary tuberculosis patients in Taiwan. BMC Public Health. 2012;12:630. doi: 10.1186/1471-2458-12-630. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Björgvinsson T, Kertz SJ, Bigda-Peyton JS, et al. Psychometric properties of the CES-D-10 in a psychiatric sample. Assessment . 2013;20:429–36. doi: 10.1177/1073191113481998. [DOI] [PubMed] [Google Scholar]
- 20.Karamba W, Tong K, Salcher I. Cambodia Poverty Assessment: Toward a More Inclusive and Resilient Cambodia. World Bank. 2022 doi: 10.1596/38344. [DOI] [Google Scholar]
- 21.Lee C-H, Lee M-C, Lin H-H, et al. Pulmonary tuberculosis and delay in anti-tuberculous treatment are important risk factors for chronic obstructive pulmonary disease. PLoS One. 2012;7:e37978. doi: 10.1371/journal.pone.0037978. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Seo W, Kim HW, Kim JS, et al. Long term management of people with post-tuberculosis lung disease. Korean J Intern Med. 2024;39:7–24. doi: 10.3904/kjim.2023.395. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Gai X, Allwood B, Sun Y. Post-tuberculosis lung disease and chronic obstructive pulmonary disease. Chin Med J (Engl) 2023;136:1923–8. doi: 10.1097/CM9.0000000000002771. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Cupido G, Günther G. Post tuberculosis lung disease and tuberculosis sequelae: A narrative review. Indian J Tuberc. 2024;71:64–72. doi: 10.1016/j.ijtb.2023.04.001. [DOI] [PubMed] [Google Scholar]
- 25.Nightingale R, Chinoko B, Lesosky M, et al. Respiratory symptoms and lung function in patients treated for pulmonary tuberculosis in Malawi: a prospective cohort study. Thorax. 2022;77:1131–9. doi: 10.1136/thoraxjnl-2021-217190. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Chiang SS, Andronikou S, Roman Sinche B, et al. Posttuberculosis Lung Disease in Adolescents in Lima, Peru. Pediatrics. 2026;158:e2025074970. doi: 10.1542/peds.2025-074970. [DOI] [PubMed] [Google Scholar]
- 27.Ozoh OB, Ojo OO, Dania MG, et al. Impact of post-tuberculosis lung disease on health-related quality of life in patients from two tertiary hospitals in Lagos, Nigeria. Afr J Thorac Crit Care Med . 2021;27:47. doi: 10.7196/AJTCCM.2021.v27i2.135. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Banholzer N, Muula G, Mureithi F, et al. Mental, physical, and respiratory health in people with tuberculosis in Southern Africa: a multi-country cohort analysis. BMC Med. 2025;23:485. doi: 10.1186/s12916-025-04321-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Meghji J, Gregorius S, Madan J, et al. The long term effect of pulmonary tuberculosis on income and employment in a low income, urban setting. Thorax. 2021;76:387–95. doi: 10.1136/thoraxjnl-2020-215338. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Migliori GB, Centis R, D’Ambrosio L, et al. Post-tuberculosis lung disease: questions and answers. Eur Respir J. 2025;66:2500992. doi: 10.1183/13993003.00992-2025. [DOI] [PubMed] [Google Scholar]
- 31.Tuot S, Teo AKJ, Cazabon D, et al. Acceptability of active case finding with a seed-and-recruit model to improve tuberculosis case detection and linkage to treatment in Cambodia: A qualitative study. PLoS One. 2019;14:e0210919. doi: 10.1371/journal.pone.0210919. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Richard H, Ke X, Priyanka S, et al. World Health Organization; 2010. Validity and comparability of out-of- pocket health expenditure from household surveys: a review of the literature and current survey instruments.https://www.who.int/publications/m/item/validity-and-comparability-of-out-of-pocket-health-expenditure-from-household-surveys-a-review-of-the-literature-and-current-survey-instruments Available. [Google Scholar]
- 33.Teo AKJ, Luu KB, Garden F, et al. The relationship between a known diagnosis of tuberculosis and symptom reporting: implications for case detection strategies. Eur Respir J. 2025;66:2402521. doi: 10.1183/13993003.02521-2024. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
Data are available upon reasonable request.