Journal Pone 0282733

PLOS ONE
RESEARCH ARTICLE
Prevalence and associated factors of delayed

sputum smear conversion in patients treated
for smear positive pulmonary tuberculosis: A
retrospective follow up study in Sabah,
Malaysia
Linghui Amanda Khor1☯, Ulfa Nur Izzati A. Wahid1☯, Lee Lee Ling2☯, Sarah Michael
a1111111111 3 3
S. Liansim ☯, Jush’n Oon ☯, Mahendran Naidu Balakrishnan4☯, Wei Leik Ng ID5*, Ai
a1111111111 Theng Cheong6*
a1111111111
a1111111111 1 Luyang Health Clinic, Kota Kinabalu, Sabah, Malaysia, 2 Tamparuli Health Clinic, Tuaran, Sabah,
a1111111111 Malaysia, 3 Penampang Health Clinic, Penampang, Sabah, Malaysia, 4 Permai Polyclinic Sri Kepayan, Kota
Kinabalu, Sabah, Malaysia, 5 Department of Primary Care Medicine, Faculty of Medicine, Universiti Malaya,
Kuala Lumpur, Malaysia, 6 Department of Family Medicine, Faculty of Medicine and Health Sciences,
Universiti Putra Malaysia, Serdang, Selangor, Malaysia
☯ These authors contributed equally to this work.

OPEN ACCESS * wlng@ummc.edu.my (WLN); cheaitheng@upm.edu.my (ATC)
Citation: Khor LA, A. Wahid UNI, Ling LL, Liansim

SMS, Oon J, Balakrishnan MN, et al. (2023)
Prevalence and associated factors of delayed Abstract
sputum smear conversion in patients treated for
smear positive pulmonary tuberculosis: A
retrospective follow up study in Sabah, Malaysia. Introduction
PLoS ONE 18(3): e0282733. https://doi.org/
10.1371/journal.pone.0282733 Tuberculosis remains a major health problem globally and in Malaysia, particularly in the
Editor: Frederick Quinn, The University of Georgia,
state of Sabah. Delayed sputum conversion is associated with treatment failure, drug-resis-
UNITED STATES tant tuberculosis and mortality. We aimed to determine the prevalence of delayed sputum
Received: September 7, 2022
conversion among smear positive pulmonary tuberculosis (PTB) patients and its associated
factors in Sabah, Malaysia.
Accepted: February 20, 2023
Published: March 6, 2023 Methods

Copyright: © 2023 Khor et al. This is an open A retrospective follow up study on all patients newly diagnosed with smear positive pulmo-
access article distributed under the terms of the
nary tuberculosis from 2017 to 2019 was conducted at three government health clinics in
Creative Commons Attribution License, which
permits unrestricted use, distribution, and Sabah, utilizing data from a national electronic tuberculosis database and medical records.
reproduction in any medium, provided the original Descriptive statistics and binary logistic regression were applied for data analysis. The out-
author and source are credited. come of the study was the sputum conversion status at the end of the two-month intensive
Data Availability Statement: All relevant data are treatment phase with either successful conversion to smear negative or non-conversion.
within the manuscript. Further data cannot be
shared publicly because the data is owned by Results
Ministry of Health Malaysia. Access to data can be
requested from National Institutes of Health (NIH), 374 patients were included in the analysis. Our patients were generally younger than 60
Ministry of Health Malaysia (contact via email: years old with no medical illness and varying proportions of tuberculosis severity as judged
jppnih@moh.gov.my) and Medical Research and
by radiographic appearance and sputum bacillary load upon diagnosis. Foreigners consti-
Ethics Committee (MREC) Malaysia (contact via
email:mrecsec@moh.gov.my) for researchers who tuted 27.8% of our sample. 8.8% (confidence interval: 6.2–12.2) did not convert to smear
meet the criteria for access to confidential data. negative at the end of the intensive phase. Binary logistic regression showed that older
PLOS ONE | https://doi.org/10.1371/journal.pone.0282733 March 6, 2023 1 / 14

PLOS ONE Prevalence and associated factors of delayed sputum smear conversion for smear positive pulmonary tuberculosis
The data underlying the results presented in the patients �60 years old (adjusted odds ratio, AOR = 4.303), foreigners (AOR = 3.184) and
study are available from the Public Health Division, patients with higher sputum bacillary load at diagnosis [2+ (AOR = 5.061) and 3+ (AOR =
Sabah State Health deaprtment (contact via phone:
+608-8512555). The authors did not have special
4.992)] were more likely to have delayed sputum smear conversion.
access to the data and requested these data in the
same manner after obtaining approval from NIH Conclusion
Malaysia and MREC Malaysia.
The prevalence of delayed sputum conversion in our study was considerably low at 8.8%
Funding: The authors received no specific funding with age �60 years old, foreigners and higher pre-treatment sputum bacillary load associ-
for this work.
ated with delayed conversion. Healthcare providers should take note of these factors and
Competing interests: The authors have declared ensure the patients receive proper follow up treatment.
that no competing interests exist.
Introduction
Tuberculosis (TB) remains a significant health problem globally. It is estimated there were 1.6
million deaths due to tuberculosis worldwide in 2021, following an upward trend from 1.4 mil-
lion in 2019 and 1.5 million in 2020 [1]. TB is expected to rank second only to COVID-19 as
the cause of death from a single infectious agent in 2020 and 2021 [1]. The Southeast Asia
region bears the highest TB burden. In 2021, 45% of new TB cases were reported in Southeast
Asia, followed by Africa (23%), Western Pacific (18%) and smaller proportions in other
regions [1].
The main source of transmission for TB is from smear positive pulmonary TB (PTB)
patients via infective droplets from their lungs and throat [2]. Direct microscopic observation
of sputum smear for acid-fast bacilli (AFB) plays an important role in treatment monitoring
and this method is widely and easily available in developing countries such as Malaysia [3]. As
recommended by World Health Organization (WHO), sputum conversion rate (SCR) to
smear negative from a smear positive patient at the end of the intensive phase of anti-tubercu-
losis therapy (ATT) (at the end of the second month of treatment duration), is an operational
indicator of the national TB control programs’ capacity and an essentially important clinical
indicator of treatment response and disease prognosis [4]. Delayed sputum conversion,
defined as non-conversion to smear negative PTB at the end of intensive phase, is associated
with poorer outcomes, specifically treatment failure, and increased risk of drug resistance and
higher mortality [5–7]. Delayed sputum conversion also contributes to higher treatment cost
and additional burden to healthcare services, In Malaysia, treatment success rate for TB was
78% in 2020, lower than the global success rate of 86% in the same year [1]. Tackling delayed
sputum conversion is one important strategy to improve the TB treatment success rate.
In Sabah, one of the states in Borneo Malaysia, TB notification constituted 20% of all TB
notification nationwide between 2012 and 2018, despite representing only about 10% of the
Malaysian population [8]. The TB incidence rate in Sabah during that period was reported as
128 per 100,000 population, which was higher than the national incidence rate (97 per 100,000
population) [9]. Sabah also has a unique and diverse sociodemographic composition com-
pared to other states in Malaysia, comprising 42 ethnic groups, almost 200 sub-ethnic groups
and a large proportion of immigrants, both legal and illegal, from neighbouring countries such
as Philippines and Indonesia.
In this study, we aimed to determine the prevalence of delayed sputum conversion in
patients with smear positive PTB and its associated factors on the west coast of Sabah. It is use-
ful to obtain insight into the extent of delayed sputum conversion in this region with high TB
burden and unique demographic profile. Understanding the factors associated with delayed
sputum conversion could guide the development of public health policy to tackle this issue.

Methods
Study design and setting
A retrospective follow up study was conducted involving all new cases of patients with smear
positive PTB at three health clinics on the central west coast of Sabah, namely Tamparuli
Health Clinic in Tuaran, Penampang Health Clinic in Penampang and Luyang Health Clinic
in Kota Kinabalu, from 1st January 2017 to 31st December 2019.
These three clinics were purposely selected because they were the main clinic that treat
tuberculosis in three different districts in Sabah, namely Tuaran, Penampang and Kota Kina-
balu. These three clinics acted as the main treating centre for tuberculosis in their respective
district. Each TB unit in the respective clinic consisted of medical officer, medical assistant and
nurses. All patients suspected of having PTB would be required to submit at least two sputum
specimens for microscopic examination with at least one early morning specimen when possi-
ble. Sputum specimens were sent for acid-fast bacilli (AFB) smear and culture. AFB smears
were performed routinely in the clinics using Ziehl-Nielsen stain and were examined under
direct microscopy by medical laboratory technologists (MLT). MLT in health clinics are
trained in analyzing the AFB smear and routinely receive refresher courses. Patients with
smear positive PTB could be started on treatment in any of these clinics. Cases like smear neg-
ative PTB, extrapulmonary TB and drug-resistant TB such as multidrug-resistant TB
(MDR-TB) would be referred to the hospital for treatment initiation. Once diagnosed and
started on treatment, they could continue treatment in these clinics. Medication was provided
on a daily basis for intensive phase and weekly basis for maintenance phase. All relevant infor-
mation from case note was entered into MyTB system, an electronic TB clinical registry system
operated by the Ministry of Health Malaysia.
Study population
We included all patients newly diagnosed with smear positive PTB who were 18 years old and
above, and under regular follow up at the study sites until the end of the two-month intensive
phase. We included patients with drug-resistant TB as well. Patients who defaulted treatment before
the end of the two-month intensive phase were excluded from the study because sputum conver-
sion would only be monitored at the end of intensive phase. We also excluded those with dissemi-
nated tuberculosis because they were being followed up in hospitals instead of health clinics.
Data collection
We used universal sampling and extracted all cases that fulfilled our study criteria from the
MyTB database. Information that was readily available from MyTB database was the age at
diagnosis, ethnicity, nationality, gender, education level, smoking status at diagnosis, diabetes
and human immunodeficiency virus (HIV) status at diagnosis, sputum AFB load at diagnosis,
chest X-ray (CXR) severity at diagnosis, presence of MDR-TB and status of sputum conversion
at the end of the two-month intensive phase. Some information that was not available from
the MyTB database such as presence of other co-morbidities, alcohol dependence status at
diagnosis, duration of symptoms before diagnosis and number of days missing directly
observed therapy (DOT) were obtained from the patients’ manual medical records that were
kept in the respective health clinics. All data were recorded using data collection form.
Variables and operational definition

Smear-positive PTB was diagnosed in either one of the following ways (10): (i) two or more
positive sputum AFB smears, or (ii) one positive sputum AFB smear accompanied by

abnormalities in CXR suggestive of PTB (as determined by the physician), or (iii) one positive
sputum AFB smear and one positive sputum culture Mycobacterium tuberculosis.
The dependent variable was the delayed sputum conversion, defined as smear-positive PTB
with a sputum sample that remains AFB smear-positive at the end of the two-month intensive
phase of anti-tuberculous treatment.
The independent variables were the age at diagnosis, ethnicity, nationality, gender, educa-
tion level, smoking status at diagnosis, alcohol dependence status at diagnosis, presence of dia-
betes mellitus at diagnosis, co-morbidities, HIV status at diagnosis, sputum AFB load at
diagnosis, CXR severity at diagnosis, duration of symptoms before diagnosis, presence of
MDR-TB and number of days missing DOT. Alcohol dependence was based on the fulfilment
of three or more DSM-IV dependence criteria of substance use disorders within 12 months
[10]. Presence of diabetes mellitus at diagnosis was defined as patients who were diagnosed
with diabetes mellitus at the point of diagnosis of TB or were previously diagnosed by physi-
cian with two abnormal glucose results (if asymptomatic), one abnormal glucose result (if
symptomatic), or HbA1c >6.3% based on Malaysian guideline [11]. Co-morbidities included
medical conditions that were either recorded in the MyTB database or medical records in
clinic, such as hypertension, dyslipidemia, ischaemic heart disease, hepatitis B, chronic kidney
disease, stroke and peptic ulcer disease. HIV status at diagnosis referred either to patients with
pre-existing HIV, or who were newly diagnosed with upon diagnosis of TB. Sputum AFB load
was graded as scanty, 1+, 2+, and 3+ based on the number of AFB seen microscopically before
initiation of treatment (negative: no bacilli per 100 fields of observation; scanty: 1–9 bacilli per
100 fields; 1+: 10 to 99 bacilli per 100 fields; 2+: 1–10 per field; and 3+: >10 per field of obser-
vation). CXR severity at diagnosis was graded by treating physicians into minimal, moderate
or far advanced based on Malaysian guideline [12]. Duration of symptoms before diagnosis
was defined by the duration of symptoms suggestive of PTB from the time of onset to the
point of diagnosis, based on the history documented by physician in the medical records.
MDR-TB was defined as strains of TB that are resistant to at least two main first-line antituber-
culous drugs (i.e. isoniazid and rifampicin), as demonstrated in the sputum culture and sensi-
tivity result (10). Number of days missing DOT was defined as the number of days patient
missed the antituberculous medication during the intensive phase as documented in the medi-
cal records.
To maintain the quality of data, validation rules were implemented in the Microsoft Excel
sheet for data entry. Standardized vocabularies and units were used with regular sessions
among researchers to discuss any irregularities. Completed data entries were screened to detect
any irregularities.
Data analysis
Data was entered and analyzed using IBM Statistical Program for Social Sciences (SPSS) soft-
ware version 26. The categorical data were presented as frequency and percentage. For the age
variable, 60 years old was chosen as the cut-off point for analysis because Malaysia public
healthcare system defined people aged 60 years and above as older persons or senior citizens
with healthcare policies for older persons designed around this cut-off point of age. The out-
come of this study (dependent variable) was sputum conversion at the end of the 2-month
intensive phase, with either successful conversion to smear negative or non-conversion
(delayed sputum conversion). The association between the dependent and independent vari-
ables was examined by Chi-square test. The assumptions for Chi-square test were checked; all
expected frequencies were greater than 1 and at least 80% are greater than 5. The level of signif-
icance was set at p-value of less than 0.05. Significant variables in Chi-square test and variables

which might have clinical importance to predict delayed sputum conversion, based on litera-
ture review, researchers’ experiences and observations, were included in the bivariable logistic
regression. Bivariable and multivariable binary logistic regression were performed to deter-
mine the association between the independent variables with delayed sputum conversion. Var-
iables with p-value <0.25 from bivariable logistic regression were included in the
multivariable logistic regression model. Literature supports the inclusion of p-value < 0.25
into the multivariable regression analysis [13]. Adjusted odds ratio was calculated with p-value
of less than 0.05 considered statistically significant. Model fit was checked with Hosmer and
Lemeshow test and classification table.
Ethics
Ethical approval for this study was obtained from the Medical Research & Ethics Committee,
Ministry of Health Malaysia (NMRR-20-1581-53331). The data extracted from the MyTB data-
base were de-identified to protect patients’ confidentiality. Patients’ consent was not required
by the ethics committee as the study only analyzed secondary data from the database and med-
ical records.
Results
A total of 374 new cases of smear positive pulmonary tuberculosis that fulfilled the inclusion
criteria were included in this study. The sociodemographic and clinical characteristics of the
cohort were shown in Table 1. The majority of the patients were male in the age group of 18 to
59 years old. About one-third of our cohort (27.8%, n = 104) was foreigner.
The majority of our cohort had no known medical illness (71.4%, n = 267) with only 11.8%
(n = 43) had diabetes mellitus and 1.3% (n = 5) had HIV. Most patients did not smoke (55.1%,
n = 206) and consume alcohol (80.5%, n = 301). More patients had moderate CXR severity
(44.1%, n = 165), 1+ AFB load at diagnosis (37.2%, n = 139) and symptoms for a month or less
at diagnosis (55.0%, n = 206). Majority of the patients adhered to DOT (83.4%, n = 312) and
only 1.1% (n = 4) were diagnosed with drug-resistant TB, which were all MDR-TB.
Out of these 374 patients with smear positive pulmonary tuberculosis, 33 (8.8%, 95% confi-
dence interval, CI: 6.2–12.2) were non-converters at the end of the intensive phase which was
at the second month of anti-tuberculous therapy, thus classified as having delayed sputum
smear conversion.
For the bivariate analysis using Chi-square (Table 2) and simple logistic regression
(Table 3), age group and sputum AFB load at diagnosis were significantly associated with
delayed sputum smear conversion (p<0.05). Bivariate analysis and simple logistic regression
were not performed for the presence of HIV and MDR-TB as the frequency in those variables
were too small for meaningful analysis (5 or less).
Variables with significance of p<0.250 in univariate analysis were included in the multivar-
iable logistic regression such as age, nationality, education level, severity of CXR, duration of
symptoms and sputum AFB load. After controlling for all these factors, only age, nationality
and sputum AFB load at diagnosis were found to be significantly associated with delayed spu-
tum smear conversion. Older patients �60 years old (adjusted odds ratio, AOR = 4.303), for-
eigners (AOR = 3.184) and patients with higher sputum AFB load at diagnosis [2+
(AOR = 5.061) and 3+ (AOR = 4.992)] were more likely to have delayed sputum smear conver-
sion. This model fit was based on the Hosmer and Lemeshow test which showed non-signifi-
cant result (p-value = 0.735) and the percentage from the classification table (91.2% correctly
classified).

Table 1. Sociodemographic and clinical characteristics of patients smear positive pulmonary tuberculosis
(N = 374).
Sociodemographic characteristics Frequency, n (%)
Age
• 18 to 59 years old 319 (85.3%)
• 60 years old and above 55 (14.7%)
Gender
• Male 250 (66.8%)
• Female 124 (33.2%)
Nationality
• Malaysian 270 (72.2%)
• Foreigner 104 (27.8%)
Educational level
• Primary/ No formal education 98 (26.2%)
• Secondary/ Tertiary 114 (30.5%)
• Unknown 162 (43.3%)
Clinical characteristics Frequency, n (%)
Comorbidities
• No known medical illness 267 (71.4%)
• 1 comorbid 70 (18.7%)
• 2 comorbid and more 37 (9.9%)
HIV status
• Non-reactive 369 (98.7%)
• Reactive 5 (1.3%)
Presence of diabetes mellitus
• Yes 43 (11.5%)
• No 331 (88.5%)
Smoking status
• Non-smoker 206 (55.1%)
• Active smoker 105 (28.1%)
• Ex-smoker 63 (16.8%)
Alcohol dependence status
• No 301 (80.5%)
• Yes 73 (19.5%)
Severity of CXR
• Mild/ No lesion 116 (31.0%)
• Moderate 165 (44.1%)
• Severe 93 (24.9%)
Duration of symptoms before diagnosis
• 1 month or less 206 (55.0%)
• 2 to 6 months 130 (34.8%)
• 7 months and above 38 (10.2%)
Sputum AFB load at diagnosis
• Scanty 84 (22.5%)
• 1+ 139 (37.2%)
• 2+ 85 (22.7%)
• 3+ 66 (17.6%)
Presence of missed DOT
• No 312 (83.4%)
• Yes 62 (16.6%)
Presence of MDR-TB
• No 370 (98.9%)
• Yes 4 (1.1%)
https://doi.org/10.1371/journal.pone.0282733.t001
Discussion
Key findings from our study are: 1) the prevalence of delayed sputum conversion from 2017 to
2019 in our cohort is 8.8% (CI 6.2–12.2); 2) Older patients � 60 years old, foreigners and

Table 2. Chi-square test for association of sociodemographic and clinical variables with delayed sputum smear conversion.
Variables Converted (N = 341) Not Converted (N = 33) P-value
Age
<60 years old 295(92.5%) 24 (7.5%) 0.033�
�60 years old 46 (83.6%) 9 (16.4%)
Gender
Male 228 (91.2%) 22 (8.8%) 0.982
Female 113 (91.1%) 11 (8.9%)
Nationality
Malaysian 250 (92.6%) 20 (7.4%) 0.120
Foreigner 91 (87.5%) 13 (12.5%)
Educational level
Primary/ No formal education 88 (89.8%) 10 (10.2%) 0.274
Secondary/ Tertiary 101 (88.6%) 13 (11.4%)
Unknown 152 (93.8%) 10 (6.2%)
Smoking status
Non-Smoker 186 (90.3%) 20 (9.7%) 0.709
Active Smoker 96 (91.4%) 9 (8.6%)
Ex-Smoker 59 (93.7%) 4 (6.3%)
Comorbidities
No known medical illness 241 (90.3%) 26 (9.7%) 0.588
1 comorbid 65 (92.9%) 5 (7.1%)
2 comorbids or more 35 (94.6%) 2 (5.4%%)
Presence of diabetes mellitus
Yes 39 (90.7%) 4 (9.3%) 0.906
No 302 (91.2%) 2 (8.8%)
Severity of CXR
No lesion/ Mild 109 (94.0%) 7 (6.0%) 0.112
Moderate 152 (92.1%) 13 (7.9%)
Severe 80 (86.0%) 13 (14.0%)
Yes 275 (91.4%) 26 (8.6%) 0.797
No 66 (90.4%) 7 (9.6%)
Sputum AFB load
Scanty 81 (96.4%) 3 (3.6%) 0.000�
1+ 135(97.1%) 4 (2.9%)
2+ 71 (83.5%) 14 (16.5%)
3+ 54 (81.8%) 12 (18.2%)
Presence of missed DOTS
No 285 (91.3%) 27 (8.7%) 0.795
Yes 56 (90.3%) 6 (9.7%)
One month or less 192 (93.2%) 14 (6.8%) 0.102
2 to 6 months 113 (86.9%) 17 (13.1%)
7 months and above 36 (94.7%) 2 (5.3%)
Item in asterisk (� ) indicates p-value < 0.05.
patients with higher sputum AFB load at diagnosis are associated with delayed sputum
conversion.
Table 4 outlines the comparison of our prevalence data with a few recent studies in the past
five years. Our finding is consistent with another Malaysian study which was also conducted in
Sabah in different research sites, where they identified 7.2% of patients having delayed sputum
conversion from 2013 to 2018 [14]. In comparison, our prevalence of delayed sputum conversion
was not too high. Ibrahim (2022) and Bhatti (2021) reported a higher prevalence of 19.1% and
30.5% respectively in Malaysia [15, 16]. Similar large variation of prevalence data was observed
globally as observed in Table 4 (8.3% to 35%), and as reported in earlier studies, from 8% in

Table 3. Bivariable and multivariable logistic regression for association of sociodemographic and clinical variables with delayed sputum smear conversion.
Variables Crude OR 95% CI P-value AOR 95% CI P-value
Age
<60 years old ref 2.405 1.052–5.497 0.037� 4.303 1.405–13.176 0.011�
�60 years old
Gender
Male ref 1.009 0.473–2.153 0.982
Female
Nationality
Malaysian ref 1.786 0.853–3.737 0.124 3.184 1.184–8.563 0.022�
Foreigner
Education level
Primary/ no formal education ref 1.133 0.473–2.710 0.780 2.481 0.795–7.737 0.117
Secondary/Tertiary 0.579 0.232–1.445 0.242 0.694 0.238–2.023 0.504
Unknown
Comorbidities
Nil ref 0.713 0.262–1.930 0.505 0.450 0.136–1.493 0.192
1 comorbid 0.530 0.120–2.330 0.400 0.379 0.071–2.018 0.256
2 comorbid or more
Presence of DM
No ref 1.068 0.356–3.200 0.906
Yes
Smoking status
Non-smoker ref 0.872 0.382–1.988 0.744
Active smoker 0.631 0.207–1.918 0.417
Ex-smoker
No ref 1.122 0.467–2.696 0.797
Yes
Severity of CXR
Mild/no lesion ref 1.332 0.514–3.448 0.555 0.730 0.252–2.116 0.563
Moderate 2.530 0.966–6.629 0.059 0.998 0.308–3.232 0.997
Severe
1 month or less ref 2.063 0.980–4.344 0.057 1.685 0.740–3.839 0.214
2 to 6 months 0.762 0.166–3.497 0.727 0.590 0.118–2.954 0.521
7 months and above
Sputum AFB load
Scanty ref 0.800 0.175–3.665 0.774 0.665 0.138–3.197 0.665
1+ 5.324 1.470–19.283 0.011� 5.061 1.315–19.478 0.018�
2+ 6.000 1.617–22.263 0.007� 4.992 1.249–19.947 0.023�
3+
Presence of missed DOTs
No ref 1.131 0.446–2.866 0.795
Yes
OR = odds ratio, AOR = adjusted odds ratio, CI = confidence interval, Ref = reference. Items in asterisk (� ) indicates p-value < 0.05. No multicollinearity was found;
Hosmer and Lemeshow test, p-value = 0.735; The classification table is 91.2% correctly classified
Cameroon up to 30% in Cleveland [14–21]. This variation is observed despite the relatively stan-
dardized, effective first-line regime for antituberculous therapy worldwide. The different sociode-
mographic profiles may be one explanation. For example, Ibrahim (2022) focused on aboriginal
groups in Peninsular Malaysia (separated from Borneo Malaysia) where the prevalence was much
higher at 19.1% with no overlapping confidence limit with ours and Mokti’s (2021) study [14,
16]. Poor health accessibility may be a factor here in this group, leading to poor adherence and
delayed sputum conversion. Likewise, health accessibility is an issue observed globally and is an
important one to tackle as tuberculosis typically affects lesser developed countries [1].

Table 4. Comparison of prevalence of delayed sputum conversion and its associated factors in recent studies (year 2017–2022).
No Author (Year) Country Prevalence, % (Confidence Associated factors
interval)
1 Our study Sabah, Malaysia 8.8 (6.2–12.2) Age�60
Foreigner
Higher sputum bacillary load
2 Ibrahim MN Peninsular Malaysia (Aborigine 19.1 (15.7–22.9) Smoking
(2022) group) Diabetes mellitus
HIV infection
3 Mokti K (2021) Sabah, Malaysia 7.2 (6.2–8.2) Moderate to advanced CXR
Age> 60
Smoking
No DOTS supervisor
Non-Malaysian
Suburban residence
4 Gunda (2017) Rural Tanzania 8.3 (4.5–13.8) Male
Age>50
Sputum bacillary load of 3+
5 Azza (2019) Tunisia 35 (31–40) Diabetes mellitus
Smoking
Haemoptysis
Higher sputum bacillary load
6 Bhatti (2021) Penang, Malaysia (hospital patients) 30.5 (27.8–33.3) Age�50
Blue-collar jobs
Smoking
Higher sputum bacillary load Relapse and interruption in
treatment
7 Asemahagn (2021) Ethiopia 15.0 (11.0–19.6) Higher BMI
Higher sputum bacillary load HIV infection
Diabetes mellitus
Smoking
Societal stigma, delay in TB service
We found that the following variables were significantly associated with delayed sputum
conversion in our cohort: age � 60 years old, foreigner status and higher sputum AFB load at
diagnosis. Older age and higher sputum AFB load have consistently been shown to be associ-
ated with delayed conversion in previous studies [2, 14, 15, 17, 18]. Older age can lead to
poorer immune response, causing ineffective clearance of the bacilli. Delay in timely health-
seeking behaviour was also observed in older person, leading to poor progress in treatment for
tuberculosis where regular, close follow up in intensive phase is important [22]. The higher
sputum AFB load at diagnosis an indication of a heavier mycobacterial burden. More time
may be needed to clear the heavier load, be it lived or dead bacilli [23]. Higher bacillary load
was associated with poorer treatment outcome and higher mortality rate in tuberculosis [24].
Another possibility is treatment failure due to multidrug-resistant tuberculosis (MDR-TB) but
we did not find many MDR-TB when we traced the sputum culture result. Higher sputum
AFB load is something identifiable at diagnosis and reduction in bacillary load can be observed
as early as three days with effective treatment [25]. It is perhaps worth considering an earlier
and more rigorous follow up with sputum AFB smear for patients with higher bacillary load
on diagnosis. The current practice in primary care clinics in Malaysia is to follow up after two
to four weeks of treatment initiation, with sputum AFB smear being repeated then and after
two months of intensive therapy. We would be able to detect poor response to treatment early
and act accordingly.
In our study, we also included foreigners in our analysis. We found that foreigners in the
state of Sabah were more likely to have delayed sputum conversion, similar to Mokti et al. [14].

The social demography in the state of Sabah is peculiar and different from the rest of Malaysia
as they are known to have more foreigners. It has been estimated that non-citizen made up
30% of population in the state of Sabah, mainly from Indonesia and Philippines [26]. Foreign
nationality was associated with poorer treatment outcome for tuberculosis in another local
study [27]. It is not a practice among healthcare facilities to identify and deport any illegal
immigrants who came to seek treatment for infectious diseases such as tuberculosis in Malay-
sia although they may be encouraged to continue treatment in their home countries. It is
known that health-seeking behaviour is different among foreigners compared to locals, owing
to various factors such as fear of deportation, financial constraints and language barriers [28].
Foreigners may tend to present late with higher severity of disease. That in turn may contrib-
ute to poorer treatment outcomes such as delayed sputum conversion. We also take note that
this problem may be the tip of the iceberg as our study only captured foreigners who presented
to the clinics, not those who avoided seeking medical treatment in government healthcare
facilities.
We did not find a significant association with other variables that are significant in previous
studies, namely the presence of diabetes mellitus, CXR severity, education level, smoking sta-
tus, alcohol status, gap in treatment and duration of symptoms [2, 15, 19]. Of these factors, dia-
betes mellitus and smoking were more consistently captured in other studies as shown in
Table 4 [14–16, 20, 21]. For diabetes mellitus, while many studies showed that its presence was
associated with delayed sputum conversion, there were also studies which presented conflict-
ing findings where presence of diabetes mellitus was not associated with poorer treatment out-
come for TB such as delayed sputum conversion [29, 30]. Mahishale V (2017) reported a more
specific finding where poor glycemic control upon diagnosis of TB predicted poorer outcomes
such as delayed sputum conversion [31]. Shewade (2017) showed inadequate high-quality data
in their systematic review to delineate the effect of glycaemic control on TB treatment outcome
[32]. We hypothesized that our cohort of diabetic patients had better glycaemic control and
thus, not significantly associated with delayed sputum conversion.
Smoking had also been associated with more extensive lung disease, lung cavitation,
delayed sputum conversion at two months, higher default rates, treatment failures and relapses
[33]. Ex-smokers also showed poorer treatment outcomes for tuberculosis [33]. The negative
effect of smoking on TB treatment outcome would increase further if coupled with alcohol
drinking [34]. Conflicting findings were also seen for smoking, where some studies found that
smoking was not associated with negative outcome for tuberculosis [35, 36]. Bay JG (2022)
posed an interesting hypothesis where they deduced that their cohort of smoker did not have
poorer outcome due to better socioeconomic standing (employed, able to buy cigarettes, have
better nutrition, better educated) [37]. In our study, we did not find any association of smok-
ing with delayed sputum conversion. It may be possible that our cohort were mainly light or
early smokers with no significant lung damage yet. More data would be needed in our cohort
to explore the hypothesis of better socioeconomic profile as a protective factor against delayed
sputum conversion.
An interesting finding for us is that the prolonged duration of symptoms before diagnosis
of PTB was not significantly associated with delayed sputum conversion in our study. Some
studies demonstrated that patients with longer duration of symptom more than 2–3 months
had delayed conversion [38, 39]. This counter the argument that prolonged duration of symp-
toms may reflect higher mycobacterial burden due to progress of disease or prolonged expo-
sure to source [2]. Objective measurement of disease burdens such as initial sputum AFB load
and severity of CXR would be more useful in the initial assessment of patients with
tuberculosis.

Strength and limitation

Our study was a multicentre study where we included three main health clinics which were the
treatment centre for tuberculosis, spanned across three different districts in Sabah. Another
strength of our study was we extracted and verified data from patients’ medical records as well
instead of just relying on the MyTB database. Certain variables were either not available or
incomplete in MyTB database such as presence of other co-morbidities, alcohol status at diag-
nosis, duration of symptoms before diagnosis, sputum AFB load and number of days missing
DOTS.
There were several limitations of our study. One, the sample size in our study may not be
large enough to capture any significant difference. This was more evident for variables such as
HIV status and presence of MDR-TB as their numbers were too small in our study for mean-
ingful analysis. Two, our data on certain variables may not be that robust as well. While we did
investigate the co-morbidities of our cohort, the co-morbidities varied a lot in the types of dis-
ease, rendering the numbers inadequate for analysis. A more detailed look into the characteris-
tics of our patients with diabetes such as duration of disease, metabolic control including
glycaemic, lipid and obesity profile, and smoking habits might be helpful to explain the insig-
nificant results [40, 41]. Three, our outcome of sputum conversion was based on the result of
sputum AFB smear. It was recognized that standard AFB smears could not differentiate dead
bacilli from viable ones but performing routine sputum culture to demonstrate successful con-
version was not feasible in resource-limited settings such as ours.
Implications to practice and recommendations

Patients with risk factors for delayed sputum smear conversion at the completion of intensive
phase of anti-tuberculous therapy (ATT) after two months should be identified earlier to avoid
poorer disease outcomes, disease complications and possible progression to MDRTB. In this
study, older age and high initial bacillary load at diagnosis were shown to be significantly asso-
ciated with delayed sputum conversion. Patients with these identified risk factors should be
properly and closely monitored and treated, with consideration of a more frequent follow up
regime as opposed to current local practice, to ensure sputum conversion by the end of the
intensive treatment phase.
It is also pertinent to recognize that most foreigners with symptoms are likely to present ini-
tially to a private primary healthcare provider. Malaysia health care system consists of a dual
system in which the primary health care services are delivered by government health clinics
and private general practices (GPs). In and around the capital city of the state of Sabah, Kota
Kinabalu, there are approximately 300 GPs which act as a conduit for the initial recognition of
and diagnosis of PTB in the population. The early identification of patients with risk factors
for delayed sputum conversion, will help GPs to expedite the identification and referral of
patients to an appropriate tertiary care centre for treatment to reduce the risk of spread of TB
in the community.
Older person and foreigners are considered the vulnerable groups in this context. Future
health policy should be designed to support these populations in terms of their follow up and
treatment for tuberculosis to improve the rate of sputum conversion and ultimately treatment
outcome for tuberculosis.
We also recommend that the MyTB database includes more robust data on clinical profile
of patients such as the glycemic control upon diagnosis and the concurrent co-morbidities
besides diabetes mellitus. The data would be useful for further research to elucidate the risk
factors for poorer TB outcomes. Further research on the more effective follow up and treat-
ment regime for patients at higher risk of delayed sputum conversion is recommended.

Conclusion
The prevalence of delayed sputum conversion found in this study was considerably low at
8.8% (CI: 6.2–12.2). Foreigners, older person � 60 years old and patients with high sputum
AFB load (AFB 2+ and more) at diagnosis were at higher risk for delayed sputum conversion.
Healthcare providers should take note of these factors and ensure the patients receive proper
follow up treatment.
Supporting information
S1 File. Grading of pulmonary tuberculosis severity based on chest radiograph in adults.
(DOCX)
Acknowledgments
We would like to thank the Director-General of Health Malaysia for his permission to publish
this article and Academy of Family Physician Malaysia for giving us the platform to conduct
our research.
Author Contributions
Conceptualization: Linghui Amanda Khor, Ulfa Nur Izzati A. Wahid, Lee Lee Ling, Sarah
Michael S. Liansim, Jush’n Oon, Mahendran Naidu Balakrishnan, Wei Leik Ng, Ai Theng
Cheong.
Data curation: Linghui Amanda Khor, Ulfa Nur Izzati A. Wahid, Lee Lee Ling, Sarah Michael
S. Liansim, Jush’n Oon, Mahendran Naidu Balakrishnan, Wei Leik Ng, Ai Theng Cheong.
Formal analysis: Linghui Amanda Khor, Ulfa Nur Izzati A. Wahid, Lee Lee Ling, Sarah
Michael S. Liansim, Jush’n Oon, Mahendran Naidu Balakrishnan, Wei Leik Ng, Ai Theng
Cheong.
Investigation: Linghui Amanda Khor, Ulfa Nur Izzati A. Wahid, Lee Lee Ling, Sarah Michael
Methodology: Linghui Amanda Khor, Ulfa Nur Izzati A. Wahid, Lee Lee Ling, Sarah Michael
Project administration: Linghui Amanda Khor, Ulfa Nur Izzati A. Wahid, Lee Lee Ling,
Sarah Michael S. Liansim, Jush’n Oon, Mahendran Naidu Balakrishnan.
Resources: Linghui Amanda Khor, Ulfa Nur Izzati A. Wahid, Lee Lee Ling, Sarah Michael S.
Liansim, Jush’n Oon, Mahendran Naidu Balakrishnan, Ai Theng Cheong.
Software: Linghui Amanda Khor, Ulfa Nur Izzati A. Wahid, Lee Lee Ling, Sarah Michael S.
Liansim, Jush’n Oon, Mahendran Naidu Balakrishnan, Ai Theng Cheong.
Supervision: Wei Leik Ng, Ai Theng Cheong.
Writing – original draft: Linghui Amanda Khor, Ulfa Nur Izzati A. Wahid, Lee Lee Ling,
Sarah Michael S. Liansim, Jush’n Oon, Mahendran Naidu Balakrishnan, Wei Leik Ng.
Writing – review & editing: Wei Leik Ng, Ai Theng Cheong.
References
1. Global tuberculosis report 2022 [Internet]. 2022 [cited 15 Dec 2022]. Available from: https://www.who.
int/teams/global-tuberculosis-programme/tb-reports/global-tuberculosis-report-2022.

2. Mohd Anwar S A SMS, S M Saliluddin, Lim P Y. Factors delaying sputum conversion in smear positive
pulmonary tuberculosis: a systematic review. International Journal of Public Health and Clinical Sci-
ences. 2018; 5(3):56–61.
3. Abd Rahman N, Mokhtar K. Challenges of National TB Control Program Implementation: The Malay-
sian Experience. Procedia—Social and Behavioral Sciences. 2015; 172:578–84.
4. WHO guidelines on tuberculosis infection prevention and control: 2019 update. WHO Guidelines
Approved by the Guidelines Review Committee. Geneva2019.
5. Djouma FN, Noubom M, Ateudjieu J, Donfack H. Delay in sputum smear conversion and outcomes of
smear-positive tuberculosis patients: a retrospective cohort study in Bafoussam, Cameroon. BMC
Infect Dis. 2015; 15:139. https://doi.org/10.1186/s12879-015-0876-1 PMID: 25884844
6. Pefura-Yone EW, Kengne AP, Kuaban C. Non-conversion of sputum culture among patients with
smear positive pulmonary tuberculosis in Cameroon: a prospective cohort study. BMC Infectious Dis-
eases. 2014; 14(1):138. https://doi.org/10.1186/1471-2334-14-138 PMID: 24618155
7. Ukwaja KN, Oshi DC, Oshi SN, Alobu I. Profile and treatment outcome of smear-positive TB patients
who failed to smear convert after 2 months of treatment in Nigeria. Trans R Soc Trop Med Hyg. 2014;
108(7):431–8. https://doi.org/10.1093/trstmh/tru070 PMID: 24846911
8. Goroh MMD, Rajahram GS, Avoi R, Van Den Boogaard CHA, William T, Ralph AP, et al. Epidemiology
of tuberculosis in Sabah, Malaysia, 2012–2018. Infect Dis Poverty. 2020; 9(1):119. https://doi.org/10.
1186/s40249-020-00739-7 PMID: 32843089
9. Avoi R, Liaw YC. Tuberculosis Death Epidemiology and Its Associated Risk Factors in Sabah, Malaysia.
Int J Environ Res Public Health. 2021; 18(18).
10. American Psychiatric Association (APA). Diagnostic and statistical manual of mental disorders. 4th ed.
Washington, DC: APA; 2000.
11. Ministry of Health Malaysia. Clinical practice guidelines on management of type 2 diabetes mellitus. 6th
ed. Putrajaya, Malaysia: Malaysian Health Technology Assessment Section; 2020.
12. Ministry of Health Malaysia. Clinical practice guidelines on management of tuberculosis. 4th ed. Putra-
jaya, Malaysia: Malaysian Health Technology Assessment Section; 2021.
13. Zhang Z. Model building strategy for logistic regression: purposeful selection. Annals of Translational
Medicine. 2016; 4(6):111. https://doi.org/10.21037/atm.2016.02.15 PMID: 27127764
14. Mokti K, Md Isa Z, Sharip J, Abu Bakar SN, Atil A, Hayati F, et al. Predictors of delayed sputum smear
conversion among pulmonary tuberculosis patients in Kota Kinabalu, Malaysia: A retrospective cohort
study. Medicine (Baltimore). 2021; 100(31):e26841. https://doi.org/10.1097/MD.0000000000026841
PMID: 34397855
15. Bhatti Z, Khan AH, Sulaiman SAS, Laghari M, Ali I. Determining the risk factors associated with delayed
sputum conversion at the end of the intensive phase among tuberculosis patients. East Mediterr Health
J. 2021; 27(8):755–63. https://doi.org/10.26719/2021.27.8.755 PMID: 34486711
16. Ibrahim MN, Nik Husain NR, Daud A, Chinnayah T. Epidemiology and Risk Factors of Delayed Sputum
Smear Conversion in Malaysian Aborigines with Smear-Positive Pulmonary Tuberculosis. Int J Environ
Res Public Health. 2022; 19(4). https://doi.org/10.3390/ijerph19042365 PMID: 35206552
17. Caetano Mota P, Carvalho A, Valente I, Braga R, Duarte R. Predictors of delayed sputum smear and
culture conversion among a Portuguese population with pulmonary tuberculosis. Rev Port Pneumol.
2012; 18(2):72–9. https://doi.org/10.1016/j.rppneu.2011.12.005 PMID: 22277838
18. Gunda DW, Nkandala I, Kavishe GA, Kilonzo SB, Kabangila R, Mpondo BC. Prevalence and Risk Fac-
tors of Delayed Sputum Conversion among Patients Treated for Smear Positive PTB in Northwestern
Rural Tanzania: A Retrospective Cohort Study. J Trop Med. 2017; 2017:5352906. https://doi.org/10.
1155/2017/5352906 PMID: 28694828
19. Parikh R, Nataraj G, Kanade S, Khatri V, Mehta P. Time to sputum conversion in smear positive pulmo-
nary TB patients on category I DOTS and factors delaying it. J Assoc Physicians India. 2012; 60:22–6.
20. Asemahagn MA. Sputum smear conversion and associated factors among smear-positive pulmonary
tuberculosis patients in East Gojjam Zone, Northwest Ethiopia: a longitudinal study. BMC Pulmonary
Medicine. 2021; 21(1):118. https://doi.org/10.1186/s12890-021-01483-w PMID: 33832466
21. Slim A, Daghfous H, Ben Mansour A, Kchouk H, Ezzaouia A, Ben Saad S, et al. Predictive factors of
delayed sputum conversion in pulmonary tuberculosis. European Respiratory Journal. 2019; 54(suppl
63):PA4616.
22. Arora VK, Singla N, Sarin R. Profile of geriatric patients under DOTS in Revised National Tuberculosis
Control Programme. Indian J Chest Dis Allied Sci. 2003; 45(4):231–5. PMID: 12962456
23. Kang HK, Jeong BH, Lee H, Park HY, Jeon K, Huh HJ, et al. Clinical significance of smear positivity for
acid-fast bacilli after >/ = 5 months of treatment in patients with drug-susceptible pulmonary tuberculo-
sis. Medicine (Baltimore). 2016; 95(31):e4540.

24. Brahmapurkar KP, Brahmapurkar VK, Zodpey SP. Sputum smear grading and treatment outcome
among directly observed treatment-short course patients of tuberculosis unit, Jagdalpur, Bastar. J Fam-
ily Med Prim Care. 2017; 6(2):293–6. https://doi.org/10.4103/jfmpc.jfmpc_24_16 PMID: 29302535
25. Honeyborne I, McHugh TD, Phillips PP, Bannoo S, Bateson A, Carroll N, et al. Molecular bacterial load
assay, a culture-free biomarker for rapid and accurate quantification of sputum Mycobacterium tubercu-
losis bacillary load during treatment. J Clin Microbiol. 2011; 49(11):3905–11. https://doi.org/10.1128/
JCM.00547-11 PMID: 21900522
26. Hassan K. The dependency of economy of Sabah on foreign workers: a study using labour force surv-
ery, 2017. 2017 [27 Jan 2022.]. Available from: https://www.dosm.gov.my/v1/uploads/files/7_
Publication/Research_Papers/3_%20JP%20SABAH%20-%20The%20dependency%20of%
20economy%20of%20Sabah%20on%20foreign%20workers.pdf.
27. Atif M, Sulaiman SA, Shafie AA, Ali I, Asif M, Babar ZU. Treatment outcome of new smear positive pul-
monary tuberculosis patients in Penang, Malaysia. BMC Infect Dis. 2014; 14:399. https://doi.org/10.
1186/1471-2334-14-399 PMID: 25037452
28. Osman AF, Abdul Mutalib M, Tafran K, Tumin M, Chong CS. Demographic and Socioeconomic Vari-
ables Associated With Health Care-Seeking Behavior Among Foreign Workers in Malaysia. Asia Pac J
Public Health. 2020; 32(1):42–8. https://doi.org/10.1177/1010539519893801 PMID: 31924113
29. Singla R, Khan N, Al-Sharif N, Ai-Sayegh MO, Shaikh MA, Osman MM. Influence of diabetes on mani-
festations and treatment outcome of pulmonary TB patients. Int J Tuberc Lung Dis. 2006; 10(1):74–9.
PMID: 16466041
30. Singla R, Osman MM, Khan N, Al-Sharif N, Al-Sayegh MO, Shaikh MA. Factors predicting persistent
sputum smear positivity among pulmonary tuberculosis patients 2 months after treatment. Int J Tuberc
Lung Dis. 2003; 7(1):58–64. PMID: 12701836
31. Mahishale V, Avuthu S, Patil B, Lolly M, Eti A, Khan S. Effect of Poor Glycemic Control in Newly Diag-
nosed Patients with Smear-Positive Pulmonary Tuberculosis and Type-2 Diabetes Mellitus. Iran J Med
Sci. 2017; 42(2):144–51. https://doi.org/10.4103/0974-5009.115470 PMID: 28360440
32. Shewade HD, Jeyashree K, Mahajan P, Shah AN, Kirubakaran R, Rao R, et al. Effect of glycemic con-
trol and type of diabetes treatment on unsuccessful TB treatment outcomes among people with TB-Dia-
betes: A systematic review. PLoS One. 2017; 12(10):e0186697. https://doi.org/10.1371/journal.pone.
0186697 PMID: 29059214
33. Mahishale V, Patil B, Lolly M, Eti A, Khan S. Prevalence of Smoking and Its Impact on Treatment Out-
comes in Newly Diagnosed Pulmonary Tuberculosis Patients: A Hospital-Based Prospective Study.
Chonnam Med J. 2015; 51(2):86–90. https://doi.org/10.4068/cmj.2015.51.2.86 PMID: 26306303
34. Ma Y, Che NY, Liu YH, Shu W, Du J, Xie SH, et al. The joint impact of smoking plus alcohol drinking on
treatment of pulmonary tuberculosis. Eur J Clin Microbiol Infect Dis. 2019; 38(4):651–7. https://doi.org/
10.1007/s10096-019-03489-z PMID: 30771122
35. Pai M, Mohan A, Dheda K, Leung CC, Yew WW, Christopher DJ, et al. Lethal interaction: the colliding
epidemics of tobacco and tuberculosis. Expert Rev Anti Infect Ther. 2007; 5(3):385–91. https://doi.org/
10.1586/14787210.5.3.385 PMID: 17547503
36. Wang JY, Hsueh PR, Jan IS, Lee LN, Liaw YS, Yang PC, et al. The effect of smoking on tuberculosis:
different patterns and poorer outcomes. Int J Tuberc Lung Dis. 2007; 11(2):143–9. PMID: 17263283
37. Bay JG, Patsche CB, Svendsen NM, Gomes VF, Rudolf F, Wejse C. Tobacco smoking impact on tuber-
culosis treatment outcome: an observational study from West Africa. Int J Infect Dis. 2022;124 Suppl 1:
S50–S5. https://doi.org/10.1016/j.ijid.2022.07.067 PMID: 35914683
38. Fortun J, Martin-Davila P, Molina A, Navas E, Hermida JM, Cobo J, et al. Sputum conversion among
patients with pulmonary tuberculosis: are there implications for removal of respiratory isolation? J Anti-
microb Chemother. 2007; 59(4):794–8. https://doi.org/10.1093/jac/dkm025 PMID: 17392354
39. Singla R, Bharty SK, Gupta UA, Khayyam KU, Vohra V, Singla N, et al. Sputum smear positivity at two
months in previously untreated pulmonary tuberculosis patients. Int J Mycobacteriol. 2013; 2(4):199–
205. https://doi.org/10.1016/j.ijmyco.2013.08.002 PMID: 26786122
40. Magee MJ, Salindri AD, Kyaw NTT, Auld SC, Haw JS, Umpierrez GE. Stress Hyperglycemia in Patients
with Tuberculosis Disease: Epidemiology and Clinical Implications. Curr Diab Rep. 2018; 18(9):71.
https://doi.org/10.1007/s11892-018-1036-y PMID: 30090969
41. Ngo MD, Bartlett S, Ronacher K. Diabetes-Associated Susceptibility to Tuberculosis: Contribution of
Hyperglycemia vs. Dyslipidemia. Microorganisms. 2021; 9(11).

Journal Pone 0282733

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Journal Pone 0282733

Uploaded by

Copyright:

Available Formats

PLOS ONE

Prevalence and associated factors of delayed

☯ These authors contributed equally to this work.

Citation: Khor LA, A. Wahid UNI, Ling LL, Liansim

Published: March 6, 2023 Methods

PLOS ONE | https://doi.org/10.1371/journal.pone.0282733 March 6, 2023 1 / 14

PLOS ONE | https://doi.org/10.1371/journal.pone.0282733 March 6, 2023 2 / 14

Variables and operational definition

PLOS ONE | https://doi.org/10.1371/journal.pone.0282733 March 6, 2023 3 / 14

PLOS ONE | https://doi.org/10.1371/journal.pone.0282733 March 6, 2023 4 / 14

PLOS ONE | https://doi.org/10.1371/journal.pone.0282733 March 6, 2023 5 / 14

PLOS ONE | https://doi.org/10.1371/journal.pone.0282733 March 6, 2023 6 / 14

Item in asterisk (� ) indicates p-value < 0.05.

PLOS ONE | https://doi.org/10.1371/journal.pone.0282733 March 6, 2023 7 / 14

PLOS ONE | https://doi.org/10.1371/journal.pone.0282733 March 6, 2023 8 / 14

PLOS ONE | https://doi.org/10.1371/journal.pone.0282733 March 6, 2023 9 / 14

PLOS ONE | https://doi.org/10.1371/journal.pone.0282733 March 6, 2023 10 / 14

Strength and limitation

Implications to practice and recommendations

PLOS ONE | https://doi.org/10.1371/journal.pone.0282733 March 6, 2023 11 / 14

PLOS ONE | https://doi.org/10.1371/journal.pone.0282733 March 6, 2023 12 / 14

PLOS ONE | https://doi.org/10.1371/journal.pone.0282733 March 6, 2023 13 / 14

PLOS ONE | https://doi.org/10.1371/journal.pone.0282733 March 6, 2023 14 / 14

You might also like