- Email: [email protected]
Impact of integrated psycho-socio-economic support on treatment outcome in drug resistant tuberculosis – A retrospective cohort study
IJTB-282; No. of Pages 6 indian journal of tuberculosis xxx (2018) xxx–xxx
Available online at www.sciencedirect.com
ScienceDirect journal homepage: http://www.journals.elsevier.com/ indian-journal-of-tuberculosis/
Original article
Impact of integrated psycho-socio-economic support on treatment outcome in drug resistant tuberculosis – A retrospective cohort study Rama Bhatt a,*, Kamal Chopra b, Rohit Vashisht c a
District TB Officer, Ramakrishna Mission Free TB Clinic, Karol Bagh, New Delhi 110005, India Director, New Delhi Tuberculosis Centre, New Delhi 110002, India c Post-doctoral Research Fellow (Biomedical Informatics), Stanford University, Stanford 94305, USA b
article info
abstract
Article history:
Objective: To assess the impact of providing integrated psycho-socio-economic support to
Received 9 March 2018
drug resistant tuberculosis (DRTB) patients on the treatment outcome under programmatic
Accepted 28 May 2018
conditions.
Available online xxx
Study design: Retrospective cohort study. Setting: An urban district TB centre in India under the Revised National Tuberculosis Control
Keywords:
Programme.
Tuberculosis
Participants: A cohort of 123 patients who started DRTB treatment between June 2010 and
TB
May 2013.
Drug resistant TB
Methods: Patients started on treatment for DRTB between June 2010 and May 2013 who were
Psycho-socio-economic support
provided with the integrated support package for at least 3 months formed the supported
RNTCP
group while the other patients of the cohort formed the non-supported group. The treatment outcomes and sputum culture conversion rates were compared between the two groups. Results: The supported group consisted of 60 patients and the non-supported group of 63 patients. The treatment success rate was found to be significantly higher in the supported group (65% vs 46.03%; p = 0.0349). Support duration was significantly associated with lower incidence of death [HR 0.876, 95% CI 0.811–0.947; p = 0.0009] and loss to follow up [OR: 0.752, 95% CI 0.597–0.873; p = 0.0023]. The treatment failure rate was higher in the supported group (16.66% vs 4.76%) with 60% of the failures in the supported group occurring after 24 months of compliant treatment. There was no significant association found between support duration and treatment failure or sputum culture conversion. Conclusion: Integrated support seems to significantly increase the treatment success rate and improve survival and treatment adherence of DRTB patients. However, early diagnosis and effective pharmacotherapy are crucial for reducing treatment failures. © 2018 Tuberculosis Association of India. Published by Elsevier B.V. All rights reserved.
* Corresponding author. E-mail address: [email protected] (R. Bhatt). https://doi.org/10.1016/j.ijtb.2018.05.020 0019-5707/© 2018 Tuberculosis Association of India. Published by Elsevier B.V. All rights reserved.
Please cite this article in press as: Bhatt R, et al. Impact of integrated psycho-socio-economic support on treatment outcome in drug resistant tuberculosis – A retrospective cohort study, Indian J Tuberc. (2018), https://doi.org/10.1016/j.ijtb.2018.05.020
IJTB-282; No. of Pages 6
2
indian journal of tuberculosis xxx (2018) xxx–xxx
1.
Introduction
The World Health Organization (WHO) 'end TB strategy' document recognizes drug resistant TB (DRTB) to be a continuing crisis and a major challenge to end the global TB epidemic.1 Globally about 3.3% of new cases and 20% of previously treated cases are estimated to have DRTB.2 Although these numbers appear to have remained steady over the last few years, that only 41% of the estimated DRTB cases were being detected/notified,2 only 71% of these cases were being put on treatment, and only 50% achieving successful treatment outcomes, as against the Global Plan target of 75% cure rate for DRTB3 is cause for serious concern. Scaling up drug susceptibility testing (DST) services, so that DRTB cases are detected as early as possible and providing standardized drug therapy to all patients diagnosed to have DRTB are major global priorities.3 However death and default rates of 22% and 20% each reported from India4 underscore the need to develop and implement more effective drug regimens as well as adequate psycho-socio-economic support mechanisms to strengthen treatment adherence. Social protection, poverty alleviation, management of comorbidities, patient support, and action on other determinants of TB have all been listed as key elements of the three pillars of the end TB strategy1; however, objective data on the impact of such interventions is meager, poorly shared, and inadequately contextualized.5 This retrospective cohort study was undertaken to assess the impact of integrated patient support measures provided to DRTB patients at an urban district TB center in India.
2.
Objective
To retrospectively assess the impact of providing integrated psycho-socio-economic support to DRTB patients on the treatment outcome under programmatic conditions at an urban district tuberculosis center.
3.
Method
Programmatic management of DRTB (PMDT): The study was undertaken at Ramakrishna Mission Free TB Clinic, New Delhi, a designated District Tuberculosis Centre (DTC) under the Revised National Tuberculosis Control Programme (RNTCP) of India. All DRTB patients on treatment at this DTC were diagnosed at RNTCP accredited reference laboratories with standard solid or liquid mycobacterial cultures and drug sensitivity tests or rapid molecular diagnostics (Line Probe Assay or GeneXpert). Those with mycobacterial infection resistant to either rifampicin alone or to rifampicin and isoniazid were treated with RNTCP Category IV regimen for multidrug resistant (MDR) TB comprising of an intensive phase (IP) of six to nine months of kanamycin, levofloxacin, ethionamide, cycloserine, pyrazinamide, and ethambutol and a continuation phase (CP) of 18 months of levofloxacin, ethionamide, cycloserine and ethambutol. Mycobacterial infection resistant to ofloxacin and kanamycin was classified
as extensively drug resistant (XDR) TB, and was treated with RNTCP Category V regimen comprising 6–12 months (IP) of capreomycin, para-amino salicylate (PAS), moxifloxacin, highdose INH, clofazimine, linezolid, and co-amoxyclav followed by 18 months (CP) of PAS, moxifloxacin, high-dose INH, clofazimine, and linezolid. Response to treatment was monitored primarily through monthly sputum cultures from the 3rd to 7th months of treatment and then sputum cultures at threemonthly intervals from the 9th month onwards till completion of treatment. Patients who completed treatment and had been consistently culture negative (with at least 5 consecutive negative results in the last 12–15 months) were declared ‘‘cured’’. Patients were declared to have ‘‘completed treatment’’ if they had taken treatment according to RNTCP guidelines but did not meet the definition for cure or treatment failure due to lack of bacteriological results. Treatment was considered to have failed if two or more of the five cultures recorded in the final 12–15 months were positive, or if any of the final three cultures was positive.6 Integrated psycho-socio-economic support: Ramakrishna Mission Free TB Clinic has an integrated patient support system consisting of provision of nutritional supplements like milk, eggs, grain, pulses, jaggery, biscuits and cooking oil; socioeconomic support in the form of cash handouts and reimbursement of conveyance; free access to consultations with specialists, diagnostic investigations, and wherever possible therapy for other medical needs; and psychoemotional support in the form of counseling, motivation, home visits, and patient-provider group meetings. These services are tailored to the needs of the patients subject to the availability of funds and supplies. Since January 2012, following the scale up of PMDT services in Delhi, and rise in numbers of DRTB cases diagnosed and put on treatment, the integrated patient support services at the DTC were also scaled up to include DRTB patients. All patients diagnosed to have DRTB and put on second line anti tuberculous therapy were evaluated for socio-economic profile and psycho-social needs by a panel comprising the DRTB coordinator, the medical officer, the district tuberculosis officer, and the officer incharge of the Clinic. Patients judged to be in need of socioeconomic support were assigned to receive monthly cash handouts and nutritional supplements besides the other items of the integrated support package as per individual needs and availability of supplies. Study period and patient groups: This DTC was catering to a population of 7 lakh through 18 DOTS centers and 7 designated microscopy centers (DMCs) till March 2013. From April 2013, following a reorganization of DTCs to align them with the revenue districts of the state, 12 of the aforementioned DOTS centers and 4 DMCs were transferred to adjacent DTCs. Though the follow up and outcome data of the DRTB patients registered with Ramakrishna Mission Free TB Clinic prior to May 2013 continued to be reported from this DTC, the patients being treated at DOTS centers transferred to adjacent DTCs received no further integrated psycho-socio-economic support. All patients started on treatment for DRTB between June 2010 and May 2013 were eligible for inclusion in this study. Patients who received the integrated support package for at least 3 months between Jan 2012 and May 2015 comprised the
Please cite this article in press as: Bhatt R, et al. Impact of integrated psycho-socio-economic support on treatment outcome in drug resistant tuberculosis – A retrospective cohort study, Indian J Tuberc. (2018), https://doi.org/10.1016/j.ijtb.2018.05.020
IJTB-282; No. of Pages 6
3
indian journal of tuberculosis xxx (2018) xxx–xxx
supported group. The non-supported group comprised of the remaining patients characterized as follows: (i) registered patients who were transferred out to the adjacent DTCs before they could be put on the integrated support package, or who received the integrated support package for less than 3 months, but whose treatment follow up and outcome data continued to be reported to RNTCP by Ramakrishna Mission Free TB Clinic, and (ii) all patients who were not considered for the integrated support package or who could not be given the support package for at least 3 months. By May 2015 all patients included in the study had been assigned a definitive treatment outcome. The treatment records of the study patients were examined for demographic details, average reported per capita family income, body mass index (BMI), presence of co-morbidities, treatment adherence, sputum culture conversion at 6 and 12 months, and treatment outcome. Statistical analysis: Welch's t-test was used to compare the two groups and two proportion z-test for comparing sample proportions. Only 2 data points (initial BMI and family size for 1 patient each) were missing at random. These data were imputed with appropriate medians. Patients in the two groups were compared and hazard ratios (HR) were computed by utilizing Kaplan–Meier (KM) analysis followed by Cox proportional (CoxPH) regression with time-to-death as outcome of interest. The patients were right censored if (i) they were lost to follow up or (ii) suffered treatment failure. The patients who experienced treatment failure were censored because further follow up of these patients was beyond the scope of this study. Logistic regression was used to compare treatment success in the two groups vis-à-vis treatment failure and loss to follow up. The Cox proportional regression and logistic regression models were adjusted for age, gender, support duration, family size and initial BMI. Average per capita monthly income was not included in the regression models as the veracity of the reported data could not be objectively verified. The data was analyzed using R (version 3.4.3 – ‘‘Kite-Eating Tree’’).
4.
Results
Sixty patients received the integrated support package for at least 3 months during the study period. These patients availed themselves of the integrated support package for an average of 14.5 months (range 3–28; SD 6.74). 63 other patients who were registered for DRTB treatment during the same period but did not receive the integrated support package for at least 3
months formed the non-supported group. One patient who chose not to be treated despite being registered for DRTB at the center, 1 patient who was transferred out of the state while still on treatment and whose final treatment outcome could not be traced, and 1 patient who died in a hospital and whose date of death could not be traced were excluded from the study. There were no other exclusions. Group characteristics: Group characteristics are illustrated in Table 1 below. Patients in the supported and non-supported groups had statistically similar age ( p = 0.1487). The proportion of female patients was less in the supported group (41.67% vs 49.2%) but the difference was not statistically significant ( p = 0.4009). The family size of patients in both groups was statistically indistinguishable ( p = 0.7187). The average reported per capita monthly family income of the supported group (mean Rs 1121.5, SD 591.02) was significantly lower than that of the non-supported group (Rs 1667.8, 1204.08; p = 0.0027). The average initial BMI of the supported group (mean 15.44, SD 3.23) was also significantly lower than the non-supported group (17.18, 3.93; p = 0.0083). The supported group also reported more comorbidities than the non-supported group; these included diabetes mellitus (5 vs 3) and HIV co-infection (3 vs 0). The mean delay in starting aid for the supported group was 4.25 months (range 0–18; SD 4.45). In sum, the two groups were similar in terms of age, gender and family size while the supported group was worse in terms of BMI and per capita monthly income. Outcome: Table 2 illustrates the treatment outcome details of the patients in the two study groups. The supported patients had a better overall treatment success with cure/treatment completion rates improving by nearly 19 percentage points (65% vs 46.03%) and death rates reduced by more than 50% (13.33% vs 25.4%). There was an 80% reduction in loss to follow up for the supported group (5% vs 23.8%). The number of patients who failed treatment or had to be switched to the Category V regimen (for XDR TB, after being started on the Category IV regimen for MDR TB) was however higher in the group receiving integrated support (16.66% vs 4.76%). Around 60% of these failures in the supported group occurred after completion of 24 months of compliant drug therapy. The success rate in the supported group was significantly higher than in the non-supported group ( p = 0.0349). Detailed analyses to show the impact of support on the three unsuccessful outcomes viz. death, loss to follow up and treatment failure are presented below. Impact on survival: The KM curves for time-to-death of patients in the two groups is illustrated in Fig. 1.
Table 1 – Group characteristics. Variable Patients Gender (Male:female)
Supported group 60 35:25
Mean (SD) Age Initial BMI Family size Average per capita monthly income in INR Support duration (months)
28.68 (13.04) 15.44 (3.23) 5 (1.82) 1121.5 (591.02) 14.5 (6.74)
Non-supported group
p-value
63 32:31
n/a 0.4009
Mean (SD) 32.3 (14.55) 17.18 (3.93) 4.87 (2.09) 1667.8 (1204.08) n/a
p-value 0.1487 0.0083 0.7187 0.0027 n/a
Please cite this article in press as: Bhatt R, et al. Impact of integrated psycho-socio-economic support on treatment outcome in drug resistant tuberculosis – A retrospective cohort study, Indian J Tuberc. (2018), https://doi.org/10.1016/j.ijtb.2018.05.020
IJTB-282; No. of Pages 6
4
indian journal of tuberculosis xxx (2018) xxx–xxx
Table 2 – Treatment outcome of DRTB patients. Outcome status
No. of patients and (%)
Treatment success (cured + treatment complete) Died Lost to follow-up Treatment failure/switched to XDR Total
Supported
Non-supported
39 (65.00%) 8 (13.33%) 3 (5.00%) 10 (16.66%) 60
29 (46.03%) 16 (25.4%) 15 (23.8%) 3 (4.76%) 63
Fig. 1 – Kaplan Meier plots representing two groups of patients (supported and not-supported). Y-axis represents the probability of event free survival, where event is defined as death. X-axis represents time in number of days. The table represents the number of patients at a given time point.
A non-parametric log rank test revealed a significant difference between the two groups of patients with respect to time-to-death ( p = 0.04). Patients in the supported group had significantly better survival probabilities. The duration of support had a highly significant inverse association with death [HR 0.876, 95% CI 0.811–0.947; p = 0.0009] as did initial BMI [HR 0.824, 95% CI 0.719–0.944; p = 0.0052]. Males were significantly associated with higher incidence of death [HR 4.067, 95% CI 1.472–11.241; p = 0.0068]; but there was no significant association of age or family size with death (see Table 3).
Impact on loss to follow-up: The results of logistic regression analysis for treatment success vis-a-vis loss to follow up are illustrated in Table 4. The duration of support was significantly associated with lower loss to follow up [OR 0.752, 95% CI 0.597– 0.873; p = 0.0023]. Initial BMI was also significantly associated with lower loss to follow up [OR: 0.786, 95% CI 0.605–0.97; p = 0.0434], while males were more likely to be lost to follow up [OR: 6.484, 95% CI 1.401–0.6; p = 0.0264] despite being provided with the support. There was no significant difference observed in the loss to follow up between the two groups with respect to family size and age.
Table 3 – Factors affecting survival of patients – Cox regression for survival of patients.
Table 4 – Factors impacting loss to follow up (vis-à-vis treatment success).
Variable
Variable
Age Gender – male Support duration Family size Initial BMI
Hazard ratio
95% CI
p-value
1.007 4.067 0.876 0.907 0.824
0.98–1.035 1.472–11.241 0.811–0.947 0.722–1.14 0.719–0.944
0.6339 0.0068 0.0009 0.4037 0.0052
Age Gender – male Support duration Family size Initial BMI
Odds ratio
95% CI
p-value
1.065 6.484 0.752 1.188 0.786
1.017–1.125 1.401–40.274 0.597–0.873 0.869–1.679 0.605–0.97
0.0123 0.0264 0.0023 0.2908 0.0434
Please cite this article in press as: Bhatt R, et al. Impact of integrated psycho-socio-economic support on treatment outcome in drug resistant tuberculosis – A retrospective cohort study, Indian J Tuberc. (2018), https://doi.org/10.1016/j.ijtb.2018.05.020
IJTB-282; No. of Pages 6
5
indian journal of tuberculosis xxx (2018) xxx–xxx
Table 5 – Factors impacting treatment failure (vis-à-vis treatment success). Variable Age Gender – male Support duration Family size Initial BMI
Odds ratio
95% CI
p-value
1.019 2.561 0.926 0.845 0.943
0.962–1.099 0.635–12.24 0.851–0.998 0.581–1.2 0.776–1.146
0.5631 0.2036 0.0514 0.3519 0.5470
Impact on treatment failure: The results from logistic regression analysis for treatment success vis-a-vis treatment failure are illustrated in Table 5. While increase in support duration had only a marginal effect on treatment failure [OR 0.926, 95% CI 0.851–0.998; p = 0.0514], none of the other variables studied had a significant association with treatment failure. The sputum culture conversion rate for patients alive and continuing treatment at 6 months after start of treatment was worse for the supported group (81.03% vs 86.95%; p = 0.4179) and was only marginally better at 12 months after start of treatment (96.23% vs 87.8%; p = 0.1236). The difference however was not statistically significant in either case (see Table 6).
5.
Discussion
Economic growth and poverty alleviation have been linked to reduction in the incidence of tuberculosis and the mortality associated with it, both historically and in current times.7,8 It is also widely agreed that attenuation of the social determinants of TB is essential for effective TB control.9,10 However, there are few objective reports highlighting the impact of specific socio-economic interventions in arresting TB.7 Our study documents significant improvements in cure/ treatment completion rates, loss to follow up rates and survival of TB patients provided with integrated psychosocio-economic support while on treatment for DRTB at a district TB center. This center offered integrated support to its DRTB patients comprising counseling, motivation, cash handouts, nutritional supplements, and ancillary medical aid. Objective evidence for impact of individual components of the package on various outcome parameters of TB is tenuous. A meta-analysis of microfinancing and cash transfer projects for possible impact on TB control suggests that for microfinancing measures to be effective in TB control it must be augmented with additional health, nutritional and educational support.7 A Cochrane review of nutritional supplement programs for TB patients
found the data insufficient to draw conclusions about positive impact on treatment outcomes or quality of life.11 There is some evidence that counseling can have a positive impact on treatment outcomes though its significance and relation to financial support is unclear.12 A recent systematic review and meta-analysis concluded that psycho-emotional and socioeconomic interventions are associated with beneficial effects on TB treatment outcomes. However, the quality of evidence was very low.13 Our study suggests that a comprehensive psycho-socioeconomic support mechanism is likely to add the best incremental value to treatment outcome. The absence of significant impact of integrated support on treatment failure and sputum culture conversion statistics in our study also point to a complex relationship between psycho-socioeconomic intervention measures and impact parameters of TB control that may be difficult to unravel in isolation. The supported group in our study was found to have higher treatment failure percentage; with around 60% of these failures occurring after completion of 24 months of compliant drug therapy. This suggests that psycho-socio-economic interventions can have adequate impact only when used in conjunction with effective pharmacotherapy. The large number of deaths and treatment dropouts in the first three months in the non-supported group also underline the need for early diagnosis and more effective drug therapy to be used in conjunction with integrated support mechanisms that need to be initiated early during therapy. The support mechanisms reported in this study were tailored to best suit the needs as well as treatment interests of the patients. Thus, patients were provided specialist consultations for ancillary health problems as well as for drug side effects, whenever needed. For patients with problems of substance abuse, cash handouts were transferred to other responsible family members who could support these patients. For some patients who were irregular in their treatment the support was also suspended as a disincentive. Thus, any effort at scale up of this program must take into consideration these issues to ensure reproducibility. Further studies would also be needed to assess the effect of such interventions in different population groups and in areas with different baseline cure/treatment completion rates.
6.
Conclusion
Integrated support seems to significantly increase the treatment success rate and improve survival and treatment adherence of
Table 6 – Sputum culture conversion: 6 and 12 months after start of treatment. Culture conversion status
Culture negative Culture positive Not Available Not Applicablea Total a
6 months
12 months
Non-supported
Supported
Non-supported
Supported
40 6 2 15 63
47 11 0 2 60
36 5 2 20 63
51 2 1 6 60
Not applicable for extra-pulmonary cases and for death or loss to follow up prior to month of testing.
Please cite this article in press as: Bhatt R, et al. Impact of integrated psycho-socio-economic support on treatment outcome in drug resistant tuberculosis – A retrospective cohort study, Indian J Tuberc. (2018), https://doi.org/10.1016/j.ijtb.2018.05.020
IJTB-282; No. of Pages 6
6
indian journal of tuberculosis xxx (2018) xxx–xxx
DRTB patients. However, early diagnosis and effective pharmacotherapy are crucial for reducing treatment failures.
Conflicts of interest The authors have none to declare.
Acknowledgements The authors wish to acknowledge the contributions of Dr. Shalini Puri, Medical Officer, Mr. Vijay Singh, PMDT supervisor, and Mr. Girjapati Tiwari, General Assistant (Junior), all of Ramakrishna Mission Free TB Clinic, New Delhi, Dr. Jagannathan Srinivasan and his team, Sarada Research Labs, Bangalore, staff of the medical records section and other RNTCP staff of Ramakrishna Mission Free TB Clinic in generating, compiling and verifying the study data; of Dr. Nigam Shah, Associate Professor of Medicine (Biomedical Informatics), Stanford University, Stanford, Alejandro Schuler, PhD candidate, Center for Biomedical Informatics Research, Stanford University School of Medicine, Stanford, and Ms. Sandhya Kunnatur, Ramakrishna Mission Free TB Clinic, New Delhi, for data analysis; and of Swami Satyaswarupananda, Incharge, Ramakrishna Mission Free TB Clinic, New Delhi, in writing the text of this paper.
references
1. The End TB Strategy. Geneva: WHO; 2015:3. 2. Global TB Report. Geneva: WHO; 2015:54.
3. Drug Resistant TB: Surveillance and Response. Geneva: WHO; 2014. 4. TB India 2017. Revised National Tuberculosis Control Programme, Annual Status Report. New Delhi: Directorate General of Health Services, Government of India; 2017:114. 5. van den Hof S, Collins D, Leimane I, Jaramillo E, Gebhard A. Lessons Learned from Best Practices in Psycho-Socio-Economic Support for Tuberculosis Patients. 2014. http://dx.doi.org/ 10.13140/RG.2.1.3679.0002. 6. Guidelines on Programmatic Management of Drug Resistant TB (PMDT) in India. New Delhi: Directorate General of Health Services, Government of India; 2012:40–57. 7. Boccia D, Hargreaves J, Lönnroth K, et al. Cash transfer and microfinance interventions for tuberculosis control: review of the impact evidence and policy implications. Int J Tuberc Lung Dis. 2011;15(suppl 2):S37–S49. 8. Dye C, Lonnroth K, Jaramillo E, Williams BG, Raviglione M. Trends in tuberculosis incidence and their determinants in 134 countries. Bull World Health Organ. 2009;87:683–691. 9. Hargreaves JR, Boccia D, Evans CA, et al. The social determinants of tuberculosis: from evidence to action. Am J Public Health. 2011;101(4):654–662. 10. Lönnroth K, Castro KG, Chakaya JM, et al. Tuberculosis control and elimination 2010-50: cure, care, and social development. Lancet. 2010;375(9728):1814–1829. 11. Sinclair D, Abba K, Grobler L, et al. Nutritional supplements for people being treated for active tuberculosis. Cochrane Database Syst Rev. (11):2011;(11):CD006086. 12. Baral S, Aryal Y, Bhattrai R, King R, Newell JN. The importance of providing counselling and financial support to patients receiving treatment for multi-drug-resistant TB: mixed method qualitative and pilot intervention studies. BMC Public Health. 2014;14:46. 13. van Hoorn R, Jaramillo E, Collins D, Gebhard A, van den Hof S. The effects of psycho-emotional and socio-economic support for tuberculosis patients on treatment adherence and treatment outcomes – a systematic review and metaanalysis. PLOS ONE. 2016. http://dx.doi.org/10.1371/journal. pone.0154095.
Please cite this article in press as: Bhatt R, et al. Impact of integrated psycho-socio-economic support on treatment outcome in drug resistant tuberculosis – A retrospective cohort study, Indian J Tuberc. (2018), https://doi.org/10.1016/j.ijtb.2018.05.020
Available online at www.sciencedirect.com
ScienceDirect journal homepage: http://www.journals.elsevier.com/ indian-journal-of-tuberculosis/
Original article
Impact of integrated psycho-socio-economic support on treatment outcome in drug resistant tuberculosis – A retrospective cohort study Rama Bhatt a,*, Kamal Chopra b, Rohit Vashisht c a
District TB Officer, Ramakrishna Mission Free TB Clinic, Karol Bagh, New Delhi 110005, India Director, New Delhi Tuberculosis Centre, New Delhi 110002, India c Post-doctoral Research Fellow (Biomedical Informatics), Stanford University, Stanford 94305, USA b
article info
abstract
Article history:
Objective: To assess the impact of providing integrated psycho-socio-economic support to
Received 9 March 2018
drug resistant tuberculosis (DRTB) patients on the treatment outcome under programmatic
Accepted 28 May 2018
conditions.
Available online xxx
Study design: Retrospective cohort study. Setting: An urban district TB centre in India under the Revised National Tuberculosis Control
Keywords:
Programme.
Tuberculosis
Participants: A cohort of 123 patients who started DRTB treatment between June 2010 and
TB
May 2013.
Drug resistant TB
Methods: Patients started on treatment for DRTB between June 2010 and May 2013 who were
Psycho-socio-economic support
provided with the integrated support package for at least 3 months formed the supported
RNTCP
group while the other patients of the cohort formed the non-supported group. The treatment outcomes and sputum culture conversion rates were compared between the two groups. Results: The supported group consisted of 60 patients and the non-supported group of 63 patients. The treatment success rate was found to be significantly higher in the supported group (65% vs 46.03%; p = 0.0349). Support duration was significantly associated with lower incidence of death [HR 0.876, 95% CI 0.811–0.947; p = 0.0009] and loss to follow up [OR: 0.752, 95% CI 0.597–0.873; p = 0.0023]. The treatment failure rate was higher in the supported group (16.66% vs 4.76%) with 60% of the failures in the supported group occurring after 24 months of compliant treatment. There was no significant association found between support duration and treatment failure or sputum culture conversion. Conclusion: Integrated support seems to significantly increase the treatment success rate and improve survival and treatment adherence of DRTB patients. However, early diagnosis and effective pharmacotherapy are crucial for reducing treatment failures. © 2018 Tuberculosis Association of India. Published by Elsevier B.V. All rights reserved.
* Corresponding author. E-mail address: [email protected] (R. Bhatt). https://doi.org/10.1016/j.ijtb.2018.05.020 0019-5707/© 2018 Tuberculosis Association of India. Published by Elsevier B.V. All rights reserved.
Please cite this article in press as: Bhatt R, et al. Impact of integrated psycho-socio-economic support on treatment outcome in drug resistant tuberculosis – A retrospective cohort study, Indian J Tuberc. (2018), https://doi.org/10.1016/j.ijtb.2018.05.020
IJTB-282; No. of Pages 6
2
indian journal of tuberculosis xxx (2018) xxx–xxx
1.
Introduction
The World Health Organization (WHO) 'end TB strategy' document recognizes drug resistant TB (DRTB) to be a continuing crisis and a major challenge to end the global TB epidemic.1 Globally about 3.3% of new cases and 20% of previously treated cases are estimated to have DRTB.2 Although these numbers appear to have remained steady over the last few years, that only 41% of the estimated DRTB cases were being detected/notified,2 only 71% of these cases were being put on treatment, and only 50% achieving successful treatment outcomes, as against the Global Plan target of 75% cure rate for DRTB3 is cause for serious concern. Scaling up drug susceptibility testing (DST) services, so that DRTB cases are detected as early as possible and providing standardized drug therapy to all patients diagnosed to have DRTB are major global priorities.3 However death and default rates of 22% and 20% each reported from India4 underscore the need to develop and implement more effective drug regimens as well as adequate psycho-socio-economic support mechanisms to strengthen treatment adherence. Social protection, poverty alleviation, management of comorbidities, patient support, and action on other determinants of TB have all been listed as key elements of the three pillars of the end TB strategy1; however, objective data on the impact of such interventions is meager, poorly shared, and inadequately contextualized.5 This retrospective cohort study was undertaken to assess the impact of integrated patient support measures provided to DRTB patients at an urban district TB center in India.
2.
Objective
To retrospectively assess the impact of providing integrated psycho-socio-economic support to DRTB patients on the treatment outcome under programmatic conditions at an urban district tuberculosis center.
3.
Method
Programmatic management of DRTB (PMDT): The study was undertaken at Ramakrishna Mission Free TB Clinic, New Delhi, a designated District Tuberculosis Centre (DTC) under the Revised National Tuberculosis Control Programme (RNTCP) of India. All DRTB patients on treatment at this DTC were diagnosed at RNTCP accredited reference laboratories with standard solid or liquid mycobacterial cultures and drug sensitivity tests or rapid molecular diagnostics (Line Probe Assay or GeneXpert). Those with mycobacterial infection resistant to either rifampicin alone or to rifampicin and isoniazid were treated with RNTCP Category IV regimen for multidrug resistant (MDR) TB comprising of an intensive phase (IP) of six to nine months of kanamycin, levofloxacin, ethionamide, cycloserine, pyrazinamide, and ethambutol and a continuation phase (CP) of 18 months of levofloxacin, ethionamide, cycloserine and ethambutol. Mycobacterial infection resistant to ofloxacin and kanamycin was classified
as extensively drug resistant (XDR) TB, and was treated with RNTCP Category V regimen comprising 6–12 months (IP) of capreomycin, para-amino salicylate (PAS), moxifloxacin, highdose INH, clofazimine, linezolid, and co-amoxyclav followed by 18 months (CP) of PAS, moxifloxacin, high-dose INH, clofazimine, and linezolid. Response to treatment was monitored primarily through monthly sputum cultures from the 3rd to 7th months of treatment and then sputum cultures at threemonthly intervals from the 9th month onwards till completion of treatment. Patients who completed treatment and had been consistently culture negative (with at least 5 consecutive negative results in the last 12–15 months) were declared ‘‘cured’’. Patients were declared to have ‘‘completed treatment’’ if they had taken treatment according to RNTCP guidelines but did not meet the definition for cure or treatment failure due to lack of bacteriological results. Treatment was considered to have failed if two or more of the five cultures recorded in the final 12–15 months were positive, or if any of the final three cultures was positive.6 Integrated psycho-socio-economic support: Ramakrishna Mission Free TB Clinic has an integrated patient support system consisting of provision of nutritional supplements like milk, eggs, grain, pulses, jaggery, biscuits and cooking oil; socioeconomic support in the form of cash handouts and reimbursement of conveyance; free access to consultations with specialists, diagnostic investigations, and wherever possible therapy for other medical needs; and psychoemotional support in the form of counseling, motivation, home visits, and patient-provider group meetings. These services are tailored to the needs of the patients subject to the availability of funds and supplies. Since January 2012, following the scale up of PMDT services in Delhi, and rise in numbers of DRTB cases diagnosed and put on treatment, the integrated patient support services at the DTC were also scaled up to include DRTB patients. All patients diagnosed to have DRTB and put on second line anti tuberculous therapy were evaluated for socio-economic profile and psycho-social needs by a panel comprising the DRTB coordinator, the medical officer, the district tuberculosis officer, and the officer incharge of the Clinic. Patients judged to be in need of socioeconomic support were assigned to receive monthly cash handouts and nutritional supplements besides the other items of the integrated support package as per individual needs and availability of supplies. Study period and patient groups: This DTC was catering to a population of 7 lakh through 18 DOTS centers and 7 designated microscopy centers (DMCs) till March 2013. From April 2013, following a reorganization of DTCs to align them with the revenue districts of the state, 12 of the aforementioned DOTS centers and 4 DMCs were transferred to adjacent DTCs. Though the follow up and outcome data of the DRTB patients registered with Ramakrishna Mission Free TB Clinic prior to May 2013 continued to be reported from this DTC, the patients being treated at DOTS centers transferred to adjacent DTCs received no further integrated psycho-socio-economic support. All patients started on treatment for DRTB between June 2010 and May 2013 were eligible for inclusion in this study. Patients who received the integrated support package for at least 3 months between Jan 2012 and May 2015 comprised the
Please cite this article in press as: Bhatt R, et al. Impact of integrated psycho-socio-economic support on treatment outcome in drug resistant tuberculosis – A retrospective cohort study, Indian J Tuberc. (2018), https://doi.org/10.1016/j.ijtb.2018.05.020
IJTB-282; No. of Pages 6
3
indian journal of tuberculosis xxx (2018) xxx–xxx
supported group. The non-supported group comprised of the remaining patients characterized as follows: (i) registered patients who were transferred out to the adjacent DTCs before they could be put on the integrated support package, or who received the integrated support package for less than 3 months, but whose treatment follow up and outcome data continued to be reported to RNTCP by Ramakrishna Mission Free TB Clinic, and (ii) all patients who were not considered for the integrated support package or who could not be given the support package for at least 3 months. By May 2015 all patients included in the study had been assigned a definitive treatment outcome. The treatment records of the study patients were examined for demographic details, average reported per capita family income, body mass index (BMI), presence of co-morbidities, treatment adherence, sputum culture conversion at 6 and 12 months, and treatment outcome. Statistical analysis: Welch's t-test was used to compare the two groups and two proportion z-test for comparing sample proportions. Only 2 data points (initial BMI and family size for 1 patient each) were missing at random. These data were imputed with appropriate medians. Patients in the two groups were compared and hazard ratios (HR) were computed by utilizing Kaplan–Meier (KM) analysis followed by Cox proportional (CoxPH) regression with time-to-death as outcome of interest. The patients were right censored if (i) they were lost to follow up or (ii) suffered treatment failure. The patients who experienced treatment failure were censored because further follow up of these patients was beyond the scope of this study. Logistic regression was used to compare treatment success in the two groups vis-à-vis treatment failure and loss to follow up. The Cox proportional regression and logistic regression models were adjusted for age, gender, support duration, family size and initial BMI. Average per capita monthly income was not included in the regression models as the veracity of the reported data could not be objectively verified. The data was analyzed using R (version 3.4.3 – ‘‘Kite-Eating Tree’’).
4.
Results
Sixty patients received the integrated support package for at least 3 months during the study period. These patients availed themselves of the integrated support package for an average of 14.5 months (range 3–28; SD 6.74). 63 other patients who were registered for DRTB treatment during the same period but did not receive the integrated support package for at least 3
months formed the non-supported group. One patient who chose not to be treated despite being registered for DRTB at the center, 1 patient who was transferred out of the state while still on treatment and whose final treatment outcome could not be traced, and 1 patient who died in a hospital and whose date of death could not be traced were excluded from the study. There were no other exclusions. Group characteristics: Group characteristics are illustrated in Table 1 below. Patients in the supported and non-supported groups had statistically similar age ( p = 0.1487). The proportion of female patients was less in the supported group (41.67% vs 49.2%) but the difference was not statistically significant ( p = 0.4009). The family size of patients in both groups was statistically indistinguishable ( p = 0.7187). The average reported per capita monthly family income of the supported group (mean Rs 1121.5, SD 591.02) was significantly lower than that of the non-supported group (Rs 1667.8, 1204.08; p = 0.0027). The average initial BMI of the supported group (mean 15.44, SD 3.23) was also significantly lower than the non-supported group (17.18, 3.93; p = 0.0083). The supported group also reported more comorbidities than the non-supported group; these included diabetes mellitus (5 vs 3) and HIV co-infection (3 vs 0). The mean delay in starting aid for the supported group was 4.25 months (range 0–18; SD 4.45). In sum, the two groups were similar in terms of age, gender and family size while the supported group was worse in terms of BMI and per capita monthly income. Outcome: Table 2 illustrates the treatment outcome details of the patients in the two study groups. The supported patients had a better overall treatment success with cure/treatment completion rates improving by nearly 19 percentage points (65% vs 46.03%) and death rates reduced by more than 50% (13.33% vs 25.4%). There was an 80% reduction in loss to follow up for the supported group (5% vs 23.8%). The number of patients who failed treatment or had to be switched to the Category V regimen (for XDR TB, after being started on the Category IV regimen for MDR TB) was however higher in the group receiving integrated support (16.66% vs 4.76%). Around 60% of these failures in the supported group occurred after completion of 24 months of compliant drug therapy. The success rate in the supported group was significantly higher than in the non-supported group ( p = 0.0349). Detailed analyses to show the impact of support on the three unsuccessful outcomes viz. death, loss to follow up and treatment failure are presented below. Impact on survival: The KM curves for time-to-death of patients in the two groups is illustrated in Fig. 1.
Table 1 – Group characteristics. Variable Patients Gender (Male:female)
Supported group 60 35:25
Mean (SD) Age Initial BMI Family size Average per capita monthly income in INR Support duration (months)
28.68 (13.04) 15.44 (3.23) 5 (1.82) 1121.5 (591.02) 14.5 (6.74)
Non-supported group
p-value
63 32:31
n/a 0.4009
Mean (SD) 32.3 (14.55) 17.18 (3.93) 4.87 (2.09) 1667.8 (1204.08) n/a
p-value 0.1487 0.0083 0.7187 0.0027 n/a
Please cite this article in press as: Bhatt R, et al. Impact of integrated psycho-socio-economic support on treatment outcome in drug resistant tuberculosis – A retrospective cohort study, Indian J Tuberc. (2018), https://doi.org/10.1016/j.ijtb.2018.05.020
IJTB-282; No. of Pages 6
4
indian journal of tuberculosis xxx (2018) xxx–xxx
Table 2 – Treatment outcome of DRTB patients. Outcome status
No. of patients and (%)
Treatment success (cured + treatment complete) Died Lost to follow-up Treatment failure/switched to XDR Total
Supported
Non-supported
39 (65.00%) 8 (13.33%) 3 (5.00%) 10 (16.66%) 60
29 (46.03%) 16 (25.4%) 15 (23.8%) 3 (4.76%) 63
Fig. 1 – Kaplan Meier plots representing two groups of patients (supported and not-supported). Y-axis represents the probability of event free survival, where event is defined as death. X-axis represents time in number of days. The table represents the number of patients at a given time point.
A non-parametric log rank test revealed a significant difference between the two groups of patients with respect to time-to-death ( p = 0.04). Patients in the supported group had significantly better survival probabilities. The duration of support had a highly significant inverse association with death [HR 0.876, 95% CI 0.811–0.947; p = 0.0009] as did initial BMI [HR 0.824, 95% CI 0.719–0.944; p = 0.0052]. Males were significantly associated with higher incidence of death [HR 4.067, 95% CI 1.472–11.241; p = 0.0068]; but there was no significant association of age or family size with death (see Table 3).
Impact on loss to follow-up: The results of logistic regression analysis for treatment success vis-a-vis loss to follow up are illustrated in Table 4. The duration of support was significantly associated with lower loss to follow up [OR 0.752, 95% CI 0.597– 0.873; p = 0.0023]. Initial BMI was also significantly associated with lower loss to follow up [OR: 0.786, 95% CI 0.605–0.97; p = 0.0434], while males were more likely to be lost to follow up [OR: 6.484, 95% CI 1.401–0.6; p = 0.0264] despite being provided with the support. There was no significant difference observed in the loss to follow up between the two groups with respect to family size and age.
Table 3 – Factors affecting survival of patients – Cox regression for survival of patients.
Table 4 – Factors impacting loss to follow up (vis-à-vis treatment success).
Variable
Variable
Age Gender – male Support duration Family size Initial BMI
Hazard ratio
95% CI
p-value
1.007 4.067 0.876 0.907 0.824
0.98–1.035 1.472–11.241 0.811–0.947 0.722–1.14 0.719–0.944
0.6339 0.0068 0.0009 0.4037 0.0052
Age Gender – male Support duration Family size Initial BMI
Odds ratio
95% CI
p-value
1.065 6.484 0.752 1.188 0.786
1.017–1.125 1.401–40.274 0.597–0.873 0.869–1.679 0.605–0.97
0.0123 0.0264 0.0023 0.2908 0.0434
Please cite this article in press as: Bhatt R, et al. Impact of integrated psycho-socio-economic support on treatment outcome in drug resistant tuberculosis – A retrospective cohort study, Indian J Tuberc. (2018), https://doi.org/10.1016/j.ijtb.2018.05.020
IJTB-282; No. of Pages 6
5
indian journal of tuberculosis xxx (2018) xxx–xxx
Table 5 – Factors impacting treatment failure (vis-à-vis treatment success). Variable Age Gender – male Support duration Family size Initial BMI
Odds ratio
95% CI
p-value
1.019 2.561 0.926 0.845 0.943
0.962–1.099 0.635–12.24 0.851–0.998 0.581–1.2 0.776–1.146
0.5631 0.2036 0.0514 0.3519 0.5470
Impact on treatment failure: The results from logistic regression analysis for treatment success vis-a-vis treatment failure are illustrated in Table 5. While increase in support duration had only a marginal effect on treatment failure [OR 0.926, 95% CI 0.851–0.998; p = 0.0514], none of the other variables studied had a significant association with treatment failure. The sputum culture conversion rate for patients alive and continuing treatment at 6 months after start of treatment was worse for the supported group (81.03% vs 86.95%; p = 0.4179) and was only marginally better at 12 months after start of treatment (96.23% vs 87.8%; p = 0.1236). The difference however was not statistically significant in either case (see Table 6).
5.
Discussion
Economic growth and poverty alleviation have been linked to reduction in the incidence of tuberculosis and the mortality associated with it, both historically and in current times.7,8 It is also widely agreed that attenuation of the social determinants of TB is essential for effective TB control.9,10 However, there are few objective reports highlighting the impact of specific socio-economic interventions in arresting TB.7 Our study documents significant improvements in cure/ treatment completion rates, loss to follow up rates and survival of TB patients provided with integrated psychosocio-economic support while on treatment for DRTB at a district TB center. This center offered integrated support to its DRTB patients comprising counseling, motivation, cash handouts, nutritional supplements, and ancillary medical aid. Objective evidence for impact of individual components of the package on various outcome parameters of TB is tenuous. A meta-analysis of microfinancing and cash transfer projects for possible impact on TB control suggests that for microfinancing measures to be effective in TB control it must be augmented with additional health, nutritional and educational support.7 A Cochrane review of nutritional supplement programs for TB patients
found the data insufficient to draw conclusions about positive impact on treatment outcomes or quality of life.11 There is some evidence that counseling can have a positive impact on treatment outcomes though its significance and relation to financial support is unclear.12 A recent systematic review and meta-analysis concluded that psycho-emotional and socioeconomic interventions are associated with beneficial effects on TB treatment outcomes. However, the quality of evidence was very low.13 Our study suggests that a comprehensive psycho-socioeconomic support mechanism is likely to add the best incremental value to treatment outcome. The absence of significant impact of integrated support on treatment failure and sputum culture conversion statistics in our study also point to a complex relationship between psycho-socioeconomic intervention measures and impact parameters of TB control that may be difficult to unravel in isolation. The supported group in our study was found to have higher treatment failure percentage; with around 60% of these failures occurring after completion of 24 months of compliant drug therapy. This suggests that psycho-socio-economic interventions can have adequate impact only when used in conjunction with effective pharmacotherapy. The large number of deaths and treatment dropouts in the first three months in the non-supported group also underline the need for early diagnosis and more effective drug therapy to be used in conjunction with integrated support mechanisms that need to be initiated early during therapy. The support mechanisms reported in this study were tailored to best suit the needs as well as treatment interests of the patients. Thus, patients were provided specialist consultations for ancillary health problems as well as for drug side effects, whenever needed. For patients with problems of substance abuse, cash handouts were transferred to other responsible family members who could support these patients. For some patients who were irregular in their treatment the support was also suspended as a disincentive. Thus, any effort at scale up of this program must take into consideration these issues to ensure reproducibility. Further studies would also be needed to assess the effect of such interventions in different population groups and in areas with different baseline cure/treatment completion rates.
6.
Conclusion
Integrated support seems to significantly increase the treatment success rate and improve survival and treatment adherence of
Table 6 – Sputum culture conversion: 6 and 12 months after start of treatment. Culture conversion status
Culture negative Culture positive Not Available Not Applicablea Total a
6 months
12 months
Non-supported
Supported
Non-supported
Supported
40 6 2 15 63
47 11 0 2 60
36 5 2 20 63
51 2 1 6 60
Not applicable for extra-pulmonary cases and for death or loss to follow up prior to month of testing.
Please cite this article in press as: Bhatt R, et al. Impact of integrated psycho-socio-economic support on treatment outcome in drug resistant tuberculosis – A retrospective cohort study, Indian J Tuberc. (2018), https://doi.org/10.1016/j.ijtb.2018.05.020
IJTB-282; No. of Pages 6
6
indian journal of tuberculosis xxx (2018) xxx–xxx
DRTB patients. However, early diagnosis and effective pharmacotherapy are crucial for reducing treatment failures.
Conflicts of interest The authors have none to declare.
Acknowledgements The authors wish to acknowledge the contributions of Dr. Shalini Puri, Medical Officer, Mr. Vijay Singh, PMDT supervisor, and Mr. Girjapati Tiwari, General Assistant (Junior), all of Ramakrishna Mission Free TB Clinic, New Delhi, Dr. Jagannathan Srinivasan and his team, Sarada Research Labs, Bangalore, staff of the medical records section and other RNTCP staff of Ramakrishna Mission Free TB Clinic in generating, compiling and verifying the study data; of Dr. Nigam Shah, Associate Professor of Medicine (Biomedical Informatics), Stanford University, Stanford, Alejandro Schuler, PhD candidate, Center for Biomedical Informatics Research, Stanford University School of Medicine, Stanford, and Ms. Sandhya Kunnatur, Ramakrishna Mission Free TB Clinic, New Delhi, for data analysis; and of Swami Satyaswarupananda, Incharge, Ramakrishna Mission Free TB Clinic, New Delhi, in writing the text of this paper.
references
1. The End TB Strategy. Geneva: WHO; 2015:3. 2. Global TB Report. Geneva: WHO; 2015:54.
3. Drug Resistant TB: Surveillance and Response. Geneva: WHO; 2014. 4. TB India 2017. Revised National Tuberculosis Control Programme, Annual Status Report. New Delhi: Directorate General of Health Services, Government of India; 2017:114. 5. van den Hof S, Collins D, Leimane I, Jaramillo E, Gebhard A. Lessons Learned from Best Practices in Psycho-Socio-Economic Support for Tuberculosis Patients. 2014. http://dx.doi.org/ 10.13140/RG.2.1.3679.0002. 6. Guidelines on Programmatic Management of Drug Resistant TB (PMDT) in India. New Delhi: Directorate General of Health Services, Government of India; 2012:40–57. 7. Boccia D, Hargreaves J, Lönnroth K, et al. Cash transfer and microfinance interventions for tuberculosis control: review of the impact evidence and policy implications. Int J Tuberc Lung Dis. 2011;15(suppl 2):S37–S49. 8. Dye C, Lonnroth K, Jaramillo E, Williams BG, Raviglione M. Trends in tuberculosis incidence and their determinants in 134 countries. Bull World Health Organ. 2009;87:683–691. 9. Hargreaves JR, Boccia D, Evans CA, et al. The social determinants of tuberculosis: from evidence to action. Am J Public Health. 2011;101(4):654–662. 10. Lönnroth K, Castro KG, Chakaya JM, et al. Tuberculosis control and elimination 2010-50: cure, care, and social development. Lancet. 2010;375(9728):1814–1829. 11. Sinclair D, Abba K, Grobler L, et al. Nutritional supplements for people being treated for active tuberculosis. Cochrane Database Syst Rev. (11):2011;(11):CD006086. 12. Baral S, Aryal Y, Bhattrai R, King R, Newell JN. The importance of providing counselling and financial support to patients receiving treatment for multi-drug-resistant TB: mixed method qualitative and pilot intervention studies. BMC Public Health. 2014;14:46. 13. van Hoorn R, Jaramillo E, Collins D, Gebhard A, van den Hof S. The effects of psycho-emotional and socio-economic support for tuberculosis patients on treatment adherence and treatment outcomes – a systematic review and metaanalysis. PLOS ONE. 2016. http://dx.doi.org/10.1371/journal. pone.0154095.
Please cite this article in press as: Bhatt R, et al. Impact of integrated psycho-socio-economic support on treatment outcome in drug resistant tuberculosis – A retrospective cohort study, Indian J Tuberc. (2018), https://doi.org/10.1016/j.ijtb.2018.05.020