| Home | About | Journals | Submit | Contact Us | Français |
While Indian studies have assessed care providers’ knowledge and practices, there is no systematic review on the quality of tuberculosis (TB) care.
We searched multiple sources to identify studies (2000–2014) on providers’ knowledge and practices. We used the International Standards for TB Care to benchmark quality of care.
Of the 47 studies included, 35 were questionnaire surveys and 12 used chart abstraction. None assessed actual practice using standardised patients. Heterogeneity in the findings precluded meta-analysis. Of 22 studies evaluating provider knowledge about using sputum smears for diagnosis, 10 found that less than half of providers had correct knowledge; 3 of 4 studies assessing self-reported practices by providers found that less than a quarter reported ordering smears for patients with chest symptoms. In 11 of 14 studies that assessed treatment, less than one third of providers knew the standard regimen for drug-susceptible TB. Adherence to standards in practice was generally lower than correct knowledge of those standards. Eleven studies with both public and private providers found higher levels of appropriate knowledge/practice in the public sector.
Available evidence suggests suboptimal quality of TB care, particularly in the private sector. Improvement of quality of care should be a priority for India.
WITH AN ESTIMATED ANNUAL incidence of 2.0–2.3 million tuberculosis (TB) cases and about 150 000–350 000 deaths per year, India has the highest TB burden in the world.1 TB control depends on early diagnosis of pulmonary TB cases and their treatment with a full course of anti-tuberculosis drugs.2 For early diagnosis of TB, all persons with cough of ![[gt-or-equal, slanted]](/corehtml/pmc/pmcents/ges.gif)
2 weeks should be referred for TB evaluation.3 Correct treatment requires the use of standardised drug regimens recommended by the World Health Organization (WHO),2 the International Standards for TB Care (ISTC) and the Standards for TB Care in India (STCI).2–4
A recent systematic review showed that in India there is a delay of nearly 2 months in making a diagnosis of TB; patients are seen by on average three different providers before a diagnosis is made.5 Drug prescription analyses have shown that irrational and inappropriate anti-tuberculosis drug regimens are widely used.6 These studies suggest that the quality of TB care in India is a matter of concern.7
The Indian health care delivery landscape is complex and fragmented, with many types of care providers in the public and private sectors.8 Studies suggest that 80% of the first-contact health care and nearly 50% of TB care occurs in the private sector.9 The private sector is also very fragmented, with both unqualified (i.e., no formal medical degree) and qualified providers (a formal degree in either allopathic medicine or in alternative traditions such as Ayurveda, Unani, Siddha and Homeopathy [AYUSH]).10,11 A recent study of 100 villages in rural Madhya Pradesh found that, among primary care providers identifying themselves as ‘doctors’, 65% reported having no formal medical training, while 25% had AYUSH degrees, and only 10% reported having an MBBS (Bachelor of Medicine, Bachelor of Surgery, i.e., formal allopathic medicine) degree. The quality of medical care was highly variable, and was found to be deficient on many levels.12,13
While several studies in India have investigated different aspects of health care providers’ knowledge and practices related to TB diagnosis and treatment, this literature has not been systematically reviewed or benchmarked against international standards.
We systematically reviewed studies that provided information on both public and private sector health care providers’ knowledge and practices related to TB diagnosis and treatment compared with the second edition (2009 version) of the ISTC.14 The ISTC was used as the benchmark for three reasons: 1) the standards that make up the ISTC were developed by a team of international experts from the public and private sectors, and are recognised as defining a widely accepted level of care to which all providers should adhere; 2) the current national guidelines of India’s Revised National Tuberculosis Control Programme (RNTCP) and the recent Standards for TB Care in India (STCI)4 are largely concordant with the second edition of the ISTC (Table 1); and 3) adherence to most of the 21 standards mentioned in the ISTC can be measured using quality indicators. Although a third edition of the ISTC has recently been published,3 we used the second edition, as the third edition emphasises the use of new diagnostic modalities such as Xpert® MTB/RIF (Cepheid, Sunnyvale, CA, USA), which have not been available in India until recently.
A medical librarian searched PubMed, Embase and the Web of Science for studies published between January 2000 and Sept 2014, without any language restrictions, using search terms for ‘tuberculosis’, ‘knowledge’, ‘practice’, ‘health care providers’ and ‘India’ (see Appendix).* In addition, we carried out electronic searches of several Indian journals to increase the yield of relevant studies, especially from non-indexed journals, including the Journal of the Indian Medical Association, the Indian Journal of Tuberculosis, the Indian Journal of Community Medicine, the Indian Journal of Public Health, the Indian Journal of Medical Research, Lung India, the Indian Journal of Chest Diseases and Allied Sciences and the National Medical Journal of India. Additional studies were identified by searching the reference lists of the primary studies. Official reports, such as the RNTCP’s annual status reports or the WHO Joint Monitoring Mission Reports, are not included in the review, as they do not provide quantitative information on the knowledge, attitudes and practices of health care providers.
All study designs (cross-sectional, descriptive studies, case control studies, cohort studies and interventional studies) that used any method to assess knowledge, attitudes or practices, such as questionnaire surveys, prescription audits, vignette-based questionnaires, clinical observation, chart abstraction or ‘mystery client’/standardised patients were included. Purely qualitative studies, case reports and studies of very low quality (explained below) were excluded.
We assessed the quality of each study based on three criteria: methodology, sampling strategy and survey response rate (Tables 2 and and3).3). These criteria were adapted from the literature on various approaches to assessing the quality of medical care.13,15,16 In addition, we assessed the provider mix in each study, as studies that narrowly focus on one subset of providers (e.g., only allopathic doctors) may inadequately reflect the complexity of India’s health system. Studies that had a participation/response rate of <50% or that included <20 providers were considered to be of very low quality and were excluded from the analysis.
Citations identified by the search were independently assessed by two review authors (SS and RS) for their eligibility. Disagreements between the two reviewers were resolved by discussion or by consulting a third reviewer (PS).
Three reviewers (SS, RS and PS) independently extracted the data from each included study into a structured data extraction form. Disagreements were resolved through discussion and/or by consulting a fourth reviewer (MP). Data extracted from each of the studies included study characteristics (design, location, urban/rural setting, sample size) and type of health care providers included. For data on ISTC standards, we first extracted information on the specific ISTC standard(s) addressed in each study, and then the quantitative data (proportions and 95% confidence intervals [CIs]) on the knowledge or practice pertaining to each of the standards reported in each study. In studies where 95%CIs were not reported, we calculated these from the data provided in the manuscript.
Studies were broadly classified into those that measured knowledge and those that measured practice based on the methodology employed. Studies that had administered questionnaires and vignettes were considered to measure knowledge, while studies that used patient interviews, chart abstraction, clinical observation or standardised patients were considered to measure practice.
We tabulated the main characteristics of the included studies. Forest plots were generated for each ISTC standard for which data were available from at least five studies. A forest plot graphically displays the relative magnitude of the parameter of interest from multiple studies. Each dot represents the proportion of providers adhering to a guideline from a particular study (ranging between 0 and 1), and the lines around each dot represent the CI. Considerable heterogeneity in study methodologies precluded meta-analysis. Instead, we narratively synthesised key findings, highlighting general trends in the findings and critical deficiencies in the current literature and the methodologies used.
As shown in Figure 1, the literature search from all sources yielded 929 citations. Of these, 47 articles were included in the analysis. Three studies were excluded on the basis of very low quality. A list of excluded studies can be obtained from the authors.
Table 4 shows the characteristics of the 47 studies included.17–62 Fieldwork for all but two studies27,56 was conducted within the 2 years prior to their actual publication. Studies were conducted in 13 of the 37 states in India. Urban locations were more heavily represented, with 25 studies conducted exclusively in urban areas, 19 studies in both urban and rural areas, and three studies exclusively in rural areas; for one study this information was not available. Most studies that evaluated care in both urban and rural sites did not disaggregate data by location, precluding the assessment of urban vs. rural differences in quality of care.
Of the 47 studies, 46 were cross-sectional and one40 was an interventional study that provided information on changes in the knowledge of the health care providers pre- and post-intervention. In this review, we have used only the pre-intervention (baseline) information from this study.
Thirty-five studies used questionnaires to collect data, while three audited medical records or reviewed prescriptions. The remaining nine studies used multiple methods (a combination of questionnaire, vignette, chart abstraction and/or focus group discussions) to collect data. Twenty-eight studies collected data by interviewing health care providers, 13 by interviewing patients on the care that they had received, three by reviewing patients’ medical records or prescriptions, and three by a combination of provider interview and a review of medical records or prescriptions.
Of the 47 studies, three did not report on whether they evaluated public or private care providers. Many studies (n = 21) only included providers in the private sector, while a smaller number (n = 12) only included public sector providers. A notable subset (n = 11) studied providers in the same general location in both the public and private sectors, using the same questionnaires for the two groups. As such, this subset of studies provides direct comparisons of the quality of care delivered by the public and private sectors.
As regards the quality of the studies based on our pre-determined rating system (Tables 2 and and3),3), none of the studies used methodologies that were considered of sufficiently high quality for measuring the actual practices or behaviours of providers. Five studies were considered sufficiently high in quality for measuring provider knowledge for some ISTC standards, as they used hypothetical case scenarios (similar to vignettes) as part of their questionnaires. Twenty-six studies used high-quality sampling strategies (i.e., either random or comprehensive sampling), and the survey response rate was high or very high in 23 studies.
Only one study explicitly used the ISTC as a benchmark for quality.17 For all other studies, we extracted the data and matched them to the relevant ISTC standards. There were eight ISTC standards for which five or more studies provided data: Standard 1 (6 studies), Standard 2 (26 studies), Standard 5 (7 studies), Standard 8 (17 studies), Standard 9 (16 studies), Standard 10 (9 studies), Standard 13 (5 studies) and Standard 18 (6 studies). Results pertaining to the key standards, i.e., sputum examination for diagnosis (Standard 2), initiation of the recommended drug regimen among new TB cases (Standard 8), and patient support to ensure adherence (Standard 9), each with 10 or more studies, are presented here. The results pertaining to the remaining standards (1, 5, 10, 13, 18) are given in the Appendix.
Of the 26 studies that provided information on Standard 2 (Figure 2), 21 assessed awareness or knowledge and five assessed practices. There was considerable heterogeneity in the proportion of providers who were aware that patients with suspected pulmonary TB should undergo sputum examination, ranging from as low as 17%29 to as high as 94%.44 Five studies that provided information on practices (mostly by interviewing patients regarding provider practices) reported that, of persons with cough of 2–3 weeks’ duration, only 11%34 to 59%36 were advised to undergo sputum examination.
Of the 17 studies that provided information on Standard 8, 14 assessed knowledge and 3 assessed practices (Figure 3). For this standard, we counted any drug regimen as meeting this standard as long as it contained the correct drugs and duration of treatment (e.g., 2 months of isoniazid [INH], rifampicin [RMP], pyrazinamide and ethambutol, followed by 4 months of INH and RMP), irrespective of whether the regimen was daily or intermittent. Almost all studies reported that less than 50% of health care providers had knowledge about the correct anti-tuberculosis treatment regimen for patients with newly diagnosed pulmonary TB, or on either the correct combination of drugs or the duration of anti-tuberculosis treatment.
Studies reporting on practice had heterogeneous findings, possibly explained by the settings in which these studies were conducted. Two studies assessed practices among in-patients in tertiary care hospital settings and found very high rates of adherence to guidelines.39,41 In contrast, one study assessed the correctness of both the combination of drugs and the dosages in the out-patient setting and found that in most cases neither the dosages nor the drug combinations were in line with ISTC recommendations.42
Of the 16 studies that provided information on Standard 9, 10 reported on whether health care providers used directly observed therapy (DOT) or a supervised approach for adherence monitoring, and six reported on whether providers had appropriate knowledge of DOT or a supervised approach (Figure 4). Of studies that assessed practice, 7 of 10 studies reported that less than half of the providers used DOT or a supervised approach. Most of their TB patients received unsupervised treatment. On the other hand, of those studies that assessed knowledge, 4 of 6 studies reported that more than 90% of the providers were aware of DOT or of a supervised treatment approach. Two studies reported that younger doctors or trainees were more likely to believe in the DOT approach.43,47
Eight studies provided direct comparisons of the quality of care delivered by the public vs. the private sector for Standards 2, 8 and 9. In all studies but one,42 adherence to all ISTC standards was found to be consistently higher in the public sector (P < 0.05, Figure 5). Five studies reported that public sector providers were more likely to know that sputum smear examination is the primary test for TB (Standard 2).32,34,52,53
Only one study suggested that private providers were marginally more likely than public providers to write an appropriate prescription for drug-susceptible TB (10% vs. 5%);42 however, prescription errors by private providers, such as too few drugs in the regimen or unnecessary use of fluoroquinolones and aminoglycosides, were more frequent than among public providers. Furthermore, studies reported that public providers were more likely to report the correct combination of drugs for treating drug-susceptible TB,42,58 to use intermittent treatment as recommended by the RNTCP,42,52,58,60 and not to use streptomycin as part of the treatment regimen for new TB cases.42 As regards patient-centred approaches to TB management, including DOT or supervised therapy (Standard 9), two studies reported higher rates of supervision in the public sector.43,60
We found six studies that reported on the proportion of providers exposed to formal training on RNTCP guidelines for TB care through workshops organised by the RNTCP. Among private providers, 17–58% reported having attended an educational session on TB care,17,35,37,52,55,58 while 73–92% of government providers reported having attended such a training session;52,58 wherever the levels of training were high, awareness levels and self-reported practices were better. In addition, the only intervention-based study in this review found dramatic improvements in knowledge about multiple ISTC standards among private sector providers 1 year after educational workshops or one-to-one training sessions.40
To our knowledge, this is the first systematic review to assess health care providers’ knowledge and practices using the ISTC as the benchmark. Our systematic review on the quality of TB care in India shows major gaps in provider knowledge and practice when benchmarked against international standards. Only half of the health care providers (from both public and private sectors) were aware of the importance of suspecting TB in persons with cough of >2–3 weeks’ duration (Appendix), and two thirds knew about using sputum smear examination for persons with presumed TB. With regard to anti-tuberculosis treatment, only a third of the providers were aware of the correct regimen for patients with initial episodes of pulmonary TB, and a third reported using DOT or a supervised approach for treatment support. This lack of awareness is surprising, and may not only explain the observed diagnostic delays shown in systematic reviews,5 but may also partly explain the high levels of treatment failure and drug resistance reported in recent studies.63,64 These data emphasise the need for greater investment in strategies that facilitate effective dissemination and implementation of the ISTC and STCI.
In studies that included both public and private health care providers, adherence to ISTC standards as measured by knowledge levels was found to be higher in the public sector. This is perhaps due to the training and monitoring of public sector providers by the RNTCP and the use of standardised protocols for case finding and treatment. In contrast, little has been done to train the vast number of private sector providers, both qualified and unqualified.
Our review findings also suggest the presence of a ‘know–do’ gap (the difference between what providers ‘know’ and what they ‘do’ in reality). As compared to self-reported or observed practices, knowledge levels on appropriate treatment of TB trended towards higher rates, especially with respect to using sputum smear microscopy and DOT. The use of standardised patient studies coupled with vignettes and chart abstraction is well suited to identifying the ‘know–do’ gap, but none of the existing studies used this methodology.
Despite a thorough literature search, we may have missed some studies from India, especially if they were published in non-indexed journals. We have also not formally explored potential publication bias, as there is no statistical test for the type of data we analysed. Those studies included had their own limitations, and were mostly based on questionnaire surveys of knowledge. The quality of most studies for assessing either knowledge and/or practice was not high, and the study methodologies were diverse. Although more than 50% of TB patients in India seek care from the public sector,9 more studies focused on private sector providers (32 studies), with fewer studies focusing on public sector providers (23 studies). However, studies that assessed the difference between public and private sectors did not provide information on whether the providers were mutually exclusive. This information is useful, as public providers can work in the private sector during off-hours, nights and weekends. The studies included in the analysis were mostly from urban areas, and did not represent all regions of the country. The urban vs. rural differences in quality of care were thus not addressed. Finally, we were unable to assess whether quality of care was related to patient load or to characteristics of the health facilities, primarily because the studies did not provide any information on these aspects.
Our findings raise several issues relevant to policy. First, substantial investment is needed in training providers on national and/or international TB guidelines in both the public and the private sector. Second, given the dominance of the private sector, and the lower levels of quality than in the public sector, serious efforts need to be made to engage the private sector in TB control, and to educate and incentivise private health providers to follow national and international standards. This is particularly critical for reducing diagnostic delays, as patients often begin their pathways to care in the private sector.65
Third, there is a need to expand the availability of recommended diagnostic and treatment services across the country and create mechanisms for all health care providers, including private sector doctors, to link their patients to these services without any obstacles. It is critical to ensure that all patients have access to affordable, quality care, regardless of where they seek care.66 Fourth, monitoring health care providers’ knowledge and practice should become a part of the routine TB surveillance system so that necessary corrective steps can be undertaken and progress can be tracked.67 Using implementation research to systematically understand and identify barriers and enablers of adherence to standards would provide an opportunity for developing targeted interventions and policy shifts that could improve TB care.
Our findings also raise methodological questions about how the quality of TB care should be measured. Available studies provide a reasonable picture of provider knowledge (i.e., what they know) and, to some extent, providers’ self-reported behaviour (i.e., what they say they do); however, these studies fail to provide any information about the behaviour of providers in real life (i.e., ‘what they actually do’). None of the studies used standardised patients. Standardised patients, also known as ‘mystery clients’, are normal (non-diseased) persons from the local community who are trained to visit health care providers, present with supposed TB symptoms and seek medical advice and care, without the providers being aware that these people are actors. The standardised patients then undergo debriefing by researchers whereby they narrate the care and advice they received from the health care providers. While standardised patient studies are resource-intensive and harder to implement, such methods have been used to successfully interrogate quality of care for other medical conditions in the Indian context.13 A pilot study on standardised patients for TB care is underway in India (J Das, personal communication), and may pave the way for evidence-based decisions on this approach.
Future studies should use rigorous, vignette-based questionnaires to assess provider knowledge. Studies suggest that the assessment of both knowledge and behaviour through well-designed vignettes may reflect provider knowledge and behaviour better than chart abstraction.15,68 Studies assessing knowledge and self-reported behaviour are still helpful in that they provide upper bounds for these various quality indicators; in other words, correct knowledge about TB care is necessary for appropriate provider behaviour, although it is certainly not sufficient to ensure appropriate behaviour. As such, although the rates of adherence to ISTC standards were quite low in this study, we believe that studies of actual provider behaviour using standardised patients could show even lower rates of adherence.
In conclusion, our review suggests poor quality of TB care in India across several international standards, particularly in the private sector. Measurement and improvement of quality of care should thus be a central component of India’s new goal of universal access to quality TB care.
This study was supported by Grand Challenges Canada, Toronto, ON, Canada, and the Bill and Melinda Gates Foundation, Seattle, WA, USA (OPP1091843). SS is a recipient of fellowships from the Canadian Thoracic Society, Ottawa, ON, Canada, and the International Union Against Tuberculosis and Lung Disease, Paris, France. RS is supported by a Harvard (Cambridge, MA, USA) T32 HIV Post-doctoral Clinical Research Fellowship (NIAID AI007433). PS is supported by a UCSF (University of California, San Francisco, CA, USA) T32 Post-doctoral Clinical Research Fellowship in Pulmonary and Critical Care Medicine (NHLBI 5T32HL007185). MP is a recipient of a career award from the Fonds de recherche du Québec – Santé, Montréal, QC, Canada. None of these funding sources had any involvement in writing of the manuscript or the decision to submit it for publication. The authors have not been paid to write this article by industry or other agency. The corresponding author (MP) has full access to all the data, and has final responsibility for the decision to submit for publication.
J Das was funded in part from the Knowledge for Change Trust Fund at The World Bank. The findings, interpretations, and conclusions expressed in this paper are entirely those of the authors. They do not necessarily represent the view of the World Bank, its Executive Directors or the countries they represent.
(((((‘tuberculosis’(mesh)) OR (‘mycobacterium tuberculosis’(mesh)) OR (tuberculosis(tw)) OR (tb(tw))) AND ((‘India’(Mesh)) OR (India*(tiab)) OR (India(ad)))) AND ((((‘Health Knowledge, Attitudes, Practice’(Mesh)) OR (‘Quality of Health Care’(-Mesh)) OR (knowledge(tiab)) OR (manage*(tiab)) OR (practic*(tiab)) OR (standard*(tiab)) OR (awareness(tiab)) OR (complian*(tiab)) OR (attitude*(-tiab))) AND ((‘health personnel’(mesh)) OR (provider*(tiab)) OR (medical officer*(tiab)) OR (physician*(tiab)) OR (doctor*(tiab)) OR (clinician*(tiab)) OR (private practi*(tiab)) OR (public practi*(tiab)) OR (medical practi*(tiab)) OR (pharmacist*(tiab)) OR (nurse*(tiab)) OR (paramedic*(-tiab)) OR ((chemist(tiab) OR chemists(tiab))) OR (AYUSH(tiab)) OR (Ayurved*(tw)) OR (Unani(tiab)) OR (Siddha(tiab)) OR (Homeopath*(tiab)) OR (practitioner*(tiab)) OR (allopath*(tiab)) OR (‘internship and residency’(mesh)) OR (intern(tiab) OR interns(tiab) OR internship*(tiab)) OR (resident(-tiab) OR residents(tiab)) OR ((residency(tiab) OR residencies(tiab))) OR (medical student*(tiab)) OR (health personnel(tiab)))) OR (‘Physician’s Practice Patterns’(Mesh)) OR (‘Standard of Care’(mesh)) OR (‘Guideline Adherence’(Mesh)) OR (‘Inappropriate Prescribing’(MESH)) OR ((‘International Standards’(tiab) AND ‘Tuberculosis Care’(tiab))) OR (ISTC(-tiab)) OR (treatment practice*(tiab)) OR (diagnostic Practice*(tiab)) OR (Prescription Practice*(tiab) OR prescribing practice*(tiab))) AND ((‘2000/01/01′(PDat) : ‘2014/12/31′(PDat)))) AND ((‘2013/10/11′(Date - Entrez) : ‘3000′(Date - Entrez))))
TS = ((tuberculosis OR tb)) AND TS = (India*) AND TS = (((knowledge OR attitude* OR practi* OR quality OR manage* OR complian* OR standard* OR awareness OR attitude*) AND (health personnel OR provider* OR medical officer* OR physician* OR doctor* OR clinician* OR private practi* OR public practi* OR medical practi* OR pharmacist* OR nurse* OR paramedic* OR chemist OR chemists OR AYUSH OR Ayurved* OR Unani OR Siddha OR Homeopath* OR practitioner* OR intern OR interns OR internship* OR resident OR residents OR medical student* OR residency OR residencies)) OR (‘guideline adher*’ OR inappropriate prescri* OR standard of care OR practice pattern* OR international standards for tuberculosis care OR treatment practice* OR ISTC OR diagnostic practice* OR prescription practice* OR prescribing practice*))
Timespan: 2000–2014. Indexes: Science Citation Index Expanded, Social Sciences Citation Index, Arts & Humanities Citation Index, Conference Proceedings Citation Index- Science, Conference Proceedings Citation Index - Social Sciences & Humanities.
Database: Embase <1996 to 2014 Week 37> Search strategy:
Forest plot of studies on ISTC Standard 1 in India (awareness/suspicion of TB in persons with cough of 2–3 weeks), ISTC Standard 5 (awareness/use of a combination of chest X-ray and sputum examination for diagnosis of sputum-negative pulmonary TB) and ISTC Standard 10 (awareness/use of sputum microscopy to monitor response to treatment). ES = effect size (proportion meeting standard); CI = confidence interval; ISTC = International Standards of TB care (2nd ed); TB = tuberculosis.
Six studies provided information on this standard, and all assessed knowledge (Figure A.1). The proportion of health care providers who were aware that tuberculosis (TB) should be suspected in persons with cough of >2–3 weeks ranged from 21%28 to 81%.54 One study comparing public vs. private sectors reported that 89% of government providers knew that cough >2–3 weeks warranted sputum examination (Standard 1), as opposed to only 48% of private providers (Figure A.3).52
Of the seven studies that provided information on this standard, four reported on awareness and two51,56 reported on practice (Figure A.1). The correct knowledge for this standard ranged from as low as 4%37 to as high as 69%.38 The three studies assessing practice followed patients in government TB registers who had submitted two sputum samples that were both smear-negative. While two studies found that 39% of patients subsequently received a chest radiograph to complete the diagnostic evaluation for smear-negative TB, one study found that this to be only 5%.16,26,50 One study comparing public vs. private sectors reported that 39% of public providers could correctly cite the appropriate criteria for diagnosis of smear-negative TB (Standard 5) as compared to only 26% of private providers.58
Nine studies reported on this standard, and all assessed provider knowledge (Figure 1). Except for two studies,17,58 all other studies reported that <40% of the providers were aware that sputum smear micros-copy is required for monitoring response to treatment for smear-positive patients. The remaining providers used clinical improvement and/or chest radiography to assess response to treatment. Two studies comparing public vs. private sectors showed that public providers were more likely to order follow-up sputum smears as part of treatment monitoring (Figure A.3).52,58
Forest plot of studies in India on ISTC Standard 13 (maintenance of a written record of TB patients initiated on treatment) and ISTC Standard 18 (screening household contacts for TB). ES = effect size (proportion meeting standard); CI = confidence interval; ISTC = International Standards of TB care (2nd ed); TB = tuberculosis.
Comparison of public vs. private health care providers’ awareness/practice on ISTC Standards 1, 10 and 13 in India. ES = effect size (proportion meeting standard); CI = confidence interval; ISTC = International Standards of TB care (2nd ed); TB = tuberculosis.
Five studies assessed whether providers maintained written records of treatment (Figure A.2). All five studies reported low levels of record maintenance. In one study, it was found that none of the health care providers in their study reported having a system to maintain written records.33 Another study assessed willingness of health care providers to maintain records and found that the majority of private sector providers were not willing to keep records.30 One study that compared the public and private sectors showed that 95% of public providers reported keeping a written treatment record for patients (Standard 13) as compared to 2% of private providers (Figure A.3).52
Of the six studies, two assessed providers’ knowledge about screening household contacts, in particular children aged <6 years, and four assessed practice pertaining to screening children. The studies that assessed knowledge were both conducted among providers in the private sector, and showed very low levels (13%29 and 19%28) of screening. The practice of screening children aged <6 years was assessed in four studies, all in TB patients treated in the public sector, and the levels ranged from 14%21 to 80%.51
Included studies provided limited information on how Indian providers diagnose and manage drug-resistant TB. ISTC Standard 11 recommends culture and drug susceptibility testing (DST) for individuals with a history of previous anti-tuberculosis treatment, ongoing smear positivity after 3 months of treatment, and treatment failure or relapse. The only study evaluating this standard found that 39% of providers reported performing DST for such cases.17 Another study of patients registered with the Revised National Tuberculosis Control Programme in Mumbai and rural areas around Pune suggests that many such patients were ‘missed’ by the system: 11% of patients who had already been placed on first-line anti-tuberculosis treatment actually had a history of previous TB, which should have merited DST during the initial provider assessment.20
*The appendix is available in the online version of this article, at http://www.ingentaconnect.com/content/iuatld/ijtld/2015/00000019/00000006/art000 .....
Conflicts of interest: None of the authors has a financial interest or conflict. MP serves as a consultant to the Bill and Melinda Gates Foundation.
PubMed Central Canada is a service of the Canadian Institutes of Health Research (CIHR) working in partnership with the National Research Council's national science library in cooperation with the National Center for Biotechnology Information at the U.S. National Library of Medicine(NCBI/NLM). It includes content provided to the PubMed Central International archive by participating publishers. |