Diagnostic practices and estimated burden of tuberculosis among children admitted to 13 government hospitals in Kenya: An analysis of two years' routine clinical data

Predictors of treatment failure among patients with pulmonary tuberculosis attending public health facilities in Nairobi county

Tuberculosis (TB) is one of the infectious diseases of public health concern globally. Kenya is ranked 15th among the 22 high TB burden countries worldwide, which collectively contribute to 80% of the world's TB cases. TB Treatment failure is one of the threats to the control of TB. The research aimed at determining affordable predictors of TB treatment failure in a resource limited setting to inform policy in designing public health interventions that are best suited to the country's needs. To determine the predictors of treatment failure among patients with sputum smear positive pulmonary TB attending selected public health facilities in Nairobi Count. Data was abstracted and summarized from both patients and their medical records, focusing on socio-demographic, behavioral, and clinical exposure data. Data was collected from 4 Sub-counties, a total of 21 public health facilities with high case load of pulmonary TB were reached. Utilizing an unmatched case-control design, the study enrolled 81 patients diagnosed with TB treatment failure (cases) and 162 patients who were declared cured after completing their anti-TB treatment (controls. Strengthen contact tracing, screening, and documentation of TB treatment failure cases. Conduct further studies to elucidate the association between HIV and TB treatment failure. The factors significantly associated with treatment failure in this study encompassed prior exposure to first-line anti-Tuberculosis drugs, positive sputum smear at 2 months of treatment, and suboptimal adherence to anti-TB treatment. These findings contribute valuable insights into the identification of simple predictors of TB treatment failure such as utilizing sputum microscopy or gene expert testing at 2 months of treatment to detect individuals at risk and strengthen the implementation of DOT and TB treatment failure contact tracing protocol.

'We did not even know it was tuberculosis': a qualitative evaluation of integrating tuberculosis services into paediatric entry points in the CaP-TB programme in Cameroon and Kenya

Introduction

Paediatric tuberculosis (TB) is often undiagnosed and under-reported. The Catalysing Paediatric TB (CaP-TB) programme provided integrated and decentralised TB screening and diagnosis services through multiple paediatric care entry points. This qualitative evaluation explores acceptability of the CaP-TB programme and existing knowledge and perceptions of paediatric TB.

Methods

A descriptive qualitative study was conducted in four sites in Kenya and six sites in Cameron. 54 in-depth interviews were conducted with caregivers, community workers (CWs) and CaP-TB programme managers, and 7 focus group discussions with healthcare workers (HCWs) and CWs. Thematic analysis identified emerging recurrent themes across participants' responses. Data were coded by using MAXQDA V.12. Data were collected during March-September 2021.

Results

Caregivers were often not aware that children were at risk for TB. HCWs reported limited knowledge about paediatric TB prior to CaP-TB. Sometimes caregivers refused to have their children tested for paediatric TB, and this was often related to a lack of awareness of paediatric TB and free services, concerns about the testing procedure and treatment and fear of stigma. TB was referred to as disease of 'shame,' associated with poverty and poor hygiene. The CaP-TB programme increased HCWs knowledge about symptoms of paediatric TB and motivation to investigate children with clinical presentations consistent with possible TB. Adding screening at all entry points was perceived to be beneficial to caregivers who would not have felt comfortable bringing their child to a TB unit. HCWs also discussed the increased workload with CaP-TB, challenges with medication stock-outs and a need for additional training.

Conclusions

CaP-TB illustrated the positive impact of decentralised paediatric TB services, including addressing the awareness and knowledge gap among caregivers and HCWs. Multiple entry points increased opportunities for identification of paediatric TB and increased caregiver comfortability with their child being tested for TB.

Trial registration number

NCT03862261.

Effect of decentralising childhood tuberculosis diagnosis to primary health centre versus district hospital levels on disease detection in children from six high tuberculosis incidence countries: an operational research, pre-post intervention study

Background

Childhood tuberculosis (TB) remains underdiagnosed largely because of limited awareness and poor access to all or any of specimen collection, molecular testing, clinical evaluation, and chest radiography at low levels of care. Decentralising childhood TB diagnostics to district hospitals (DH) and primary health centres (PHC) could improve case detection.

Methods

We conducted an operational research study using a pre-post intervention cross-sectional study design in 12 DHs and 47 PHCs of 12 districts across Cambodia, Cameroon, Côte d'Ivoire, Mozambique, Sierra Leone and Uganda. The intervention included 1) a comprehensive diagnosis package at patient-level with tuberculosis screening for all sick children and young adolescents <15 years, and clinical evaluation, Xpert Ultra-testing on respiratory and stool samples, and chest radiography for children with presumptive TB, and 2) two decentralisation approaches (PHC-focused or DH-focused) to which districts were randomly allocated at country level. We collected aggregated and individual data. We compared the proportion of tuberculosis detection in children and young adolescents <15 years pre-intervention (01 August 2018-30 November 2019) versus during intervention (07 March 2020-30 September 2021), overall and by decentralisation approach. This study is registered with ClinicalTrials.gov, NCT04038632.

Findings

TB was diagnosed in 217/255,512 (0.08%) children and young adolescent <15 years attending care pre-intervention versus 411/179,581 (0.23%) during intervention, (OR: 3.59 [95% CI 1.99-6.46], p-value<0.0001; p-value = 0.055 after correcting for over-dispersion). In DH-focused districts, TB diagnosis was 80/122,570 (0.07%) versus 302/86,186 (0.35%) (OR: 4.07 [1.86-8.90]; p-value = 0.0005; p-value = 0.12 after correcting for over-dispersion); and 137/132,942 (0.10%) versus 109/93,395 (0.11%) in PHC-focused districts, respectively (OR: 2.92 [1.25-6.81; p-value = 0.013; p-value = 0.26 after correcting for over-dispersion).

Interpretation

Decentralising and strengthening childhood TB diagnosis at lower levels of care increases tuberculosis case detection but the difference was not statistically significant.

Funding source

Unitaid, Grant number 2017-15-UBx-TB-SPEED.

Cost-effectiveness and budget impact of decentralising childhood tuberculosis diagnosis in six high tuberculosis incidence countries: a mathematical modelling study

Background

The burden of childhood tuberculosis remains high globally, largely due to under-diagnosis. Decentralising childhood tuberculosis diagnosis services to lower health system levels could improve case detection, but there is little empirically based evidence on cost-effectiveness or budget impact.

Methods

In this mathematical modelling study, we assessed the cost-effectiveness and budget impact of decentralising a comprehensive diagnosis package for childhood tuberculosis to district hospitals (DH-focused) or primary health centres (PHC-focused) compared to standard of care (SOC). The project was conducted in Cambodia, Cameroon, Côte d'Ivoire, Mozambique, Sierra Leone, and Uganda between August 1st, 2018 and September 30th, 2021. A mathematical model was developed to assess the health and economic outcomes of the intervention from a health system perspective. Estimated outcomes were tuberculosis cases, deaths, disability-adjusted life years (DALYs) and incremental cost-effectiveness ratios (ICERs). We also calculated the budget impact of nationwide implementation. The TB-Speed Decentralization study is registered with ClinicalTrials.gov, NCT04038632.

Findings

For the DH-focused strategy versus SOC, ICERs ranged between $263 (Cambodia) and $342 (Côte d'Ivoire) per DALY averted. For the PHC-focused strategy versus SOC, ICERs ranged between $477 (Cambodia) and $599 (Côte d'Ivoire) per DALY averted. Results were sensitive to TB prevalence and the discount rate used. The additional costs of implementing the DH-focused strategy ranged between $12.8 M (range 10.8-16.4) (Cambodia) and $50.4 M (36.5-74.4) (Mozambique), and between $13.9 M (12.6-15.6) (Sierra Leone) and $134.6 M (127.1-143.0) (Uganda) for the PHC-focused strategy.

Interpretation

The DH-focused strategy may be cost-effective in some countries, depending on the cost-effectiveness threshold used for policy making. Either intervention would require substantial early investment.

Funding

Unitaid.

Improving TB detection among children in routine clinical care through intensified case finding in facility-based child health entry points and decentralized management: A before-and-after study in Nine Sub-Saharan African Countries

In 2022, an estimated 1.25 million children <15 years of age developed tuberculosis (TB) worldwide, but >50% remained undiagnosed or unreported. WHO recently recommended integrated and decentralized models of care as an approach to improve access to TB services for children, but evidence remains limited. The Catalyzing Paediatric TB Innovation project (CaP-TB) implemented a multi-pronged intervention to improve TB case finding in children in nine sub-Saharan African countries. The intervention introduced systematic TB screening in different facility-based child-health entry-points, decentralisation of TB diagnosis and management, improved sample collection with access to Xpert® MTB/RIF or MTB/RIF Ultra testing, and implementation of contact investigation. Pre-intervention records were compared with those during intervention to assess effect on paediatric TB cascade of care. The intervention screened 1 991 401 children <15 years of age for TB across 144 health care facilities. The monthly paediatric TB case detection rate increased significantly during intervention versus pre-intervention (+46.0%, 95% CI 36.2-55.8%; p<0.0001), with variability across countries. The increase was greater in the <5 years old compared to the 5-14 years old (+53.4%, 95% CI 35.2-71.9%; p<0.0001 versus +39.9%, 95% CI 27.6-52.2%; p<0.0001). Relative contribution of lower-tier facilities to total case detection rate increased from 37% (71.8/191.8) pre-intervention to 50% (139.9/280.2) during intervention. The majority (89.5%) of children with TB were identified through facility-based intensified case-finding and primarily accessed care through outpatient and inpatient departments. In this multi-country study implemented under real-life conditions, the implementation of integrated and decentralized interventions increased paediatric TB case detection. The increase was driven by lower-tier facilities that serve as the primary point of healthcare contact for most patients. The effect was greater in children < 5 years compared to 5-14 years old, representing an important achievement as the TB detection gap is higher in this subpopulation. (Study number NCT03948698).

Number Needed to Screen for Tuberculosis Disease Among Children: A Systematic Review

CONTEXT

Improving detection of pediatric tuberculosis (TB) is critical to reducing morbidity and mortality among children.

OBJECTIVE

We conducted a systematic review to estimate the number of children needed to screen (NNS) to detect a single case of active TB using different active case finding (ACF) screening approaches and across different settings.

DATA SOURCES

We searched 4 databases (PubMed, Embase, Scopus, and the Cochrane Library) for articles published from November 2010 to February 2020.

STUDY SELECTION

We included studies of TB ACF in children using symptom-based screening, clinical indicators, chest x-ray, and Xpert.

DATA EXTRACTION

We indirectly estimated the weighted mean NNS for a given modality, location, and population using the inverse of the weighted prevalence. We assessed risk of bias using a modified AXIS tool.

RESULTS

We screened 27 221 titles and abstracts, of which we included 31 studies of ACF in children < 15 years old. Symptom-based screening was the most common screening modality (weighted mean NNS: 257 [range, 5-undefined], 19 studies). The weighted mean NNS was lower in both inpatient (216 [18-241]) and outpatient (67 [5-undefined]) settings (107 [5-undefined]) compared with community (1117 [28-5146]) and school settings (464 [118-665]). Risk of bias was low.

LIMITATIONS

Heterogeneity in the screening modalities and populations make it difficult to draw conclusions.

CONCLUSIONS

We identified a potential opportunity to increase TB detection by screening children presenting in health care settings. Pediatric TB case finding interventions should incorporate evidence-based interventions and local contextual information in an effort to detect as many children with TB as possible.

Effect of systematic tuberculosis detection on mortality in young children with severe pneumonia in countries with high incidence of tuberculosis: a stepped-wedge cluster-randomised trial

BACKGROUND

Tuberculosis diagnosis might be delayed or missed in children with severe pneumonia because this diagnosis is usually only considered in cases of prolonged symptoms or antibiotic failure. Systematic tuberculosis detection at hospital admission could increase case detection and reduce mortality.

METHODS

We did a stepped-wedge cluster-randomised trial in 16 hospitals from six countries (Cambodia, Cameroon, Côte d'Ivoire, Mozambique, Uganda, and Zambia) with high incidence of tuberculosis. Children younger than 5 years with WHO-defined severe pneumonia received either the standard of care (control group) or standard of care plus Xpert MTB/RIF Ultra (Xpert Ultra; Cepheid, Sunnyvale, CA, USA) on nasopharyngeal aspirate and stool samples (intervention group). Clusters (hospitals) were progressively switched from control to intervention at 5-week intervals, using a computer-generated random sequence, stratified on incidence rate of tuberculosis at country level, and masked to teams until 5 weeks before switch. We assessed the effect of the intervention on primary (12-week all-cause mortality) and secondary (including tuberculosis diagnosis) outcomes, using generalised linear mixed models. The primary analysis was by intention to treat. We described outcomes in children with severe acute malnutrition in a post hoc analysis. This study is registered with ClinicalTrials.gov (NCT03831906) and the Pan African Clinical Trial Registry (PACTR202101615120643).

FINDINGS

From March 21, 2019, to March 30, 2021, we enrolled 1401 children in the control group and 1169 children in the intervention group. In the intervention group, 1140 (97·5%) children had nasopharyngeal aspirates and 942 (80·6%) had their stool collected; 24 (2·1%) had positive Xpert Ultra. At 12 weeks, 110 (7·9%) children in the control group and 91 (7·8%) children in the intervention group had died (adjusted odds ratio [OR] 0·986, 95% CI 0·597-1·630, p=0·957), and 74 (5·3%) children in the control group and 88 (7·5%) children in the intervention group had tuberculosis diagnosed (adjusted OR 1·238, 95% CI 0·696-2·202, p=0·467). In children with severe acute malnutrition, 57 (23·8%) of 240 children in the control group and 53 (17·8%) of 297 children in the intervention group died, and 36 (15·0%) of 240 children in the control group and 56 (18·9%) of 297 children in the intervention group were diagnosed with tuberculosis. The main adverse events associated with nasopharyngeal aspirates were samples with blood in 312 (27·3%) of 1147 children with nasopharyngeal aspirates attempted, dyspnoea or SpO₂ less than 95% in 134 (11·4%) of children, and transient respiratory distress or SpO₂ less than 90% in 59 (5·2%) children. There was no serious adverse event related to nasopharyngeal aspirates reported during the trial.

INTERPRETATION

Systematic molecular tuberculosis detection at hospital admission did not reduce mortality in children with severe pneumonia. High treatment and microbiological confirmation rates support more systematic use of Xpert Ultra in this group, notably in children with severe acute malnutrition.

FUNDING

Unitaid and L'Initiative.

TRANSLATION

For the French translation of the abstract see Supplementary Materials section.

Diagnostic Advances in Childhood Tuberculosis—Improving Specimen Collection and Yield of Microbiological Diagnosis for Intrathoracic Tuberculosis

There is no microbiological gold standard for childhood tuberculosis (TB) diagnosis. The paucibacillary nature of the disease, challenges in sample collection in young children, and the limitations of currently available microbiological tests restrict microbiological confirmation of intrathoracic TB to the minority of children. Recent WHO guidelines recommend the use of novel rapid molecular assays as initial diagnostic tests for TB and endorse alternative sample collection methods for children. However, the uptake of these tools in high-endemic settings remains low. In this review, we appraise historic and new microbiological tests and sample collection techniques that can be used for the diagnosis of intrathoracic TB in children. We explore challenges and possible ways to improve diagnostic yield despite limitations, and identify research gaps to address in order to improve the microbiological diagnosis of intrathoracic TB in children.

Pulse oximetry adoption and oxygen orders at paediatric admission over 7 years in Kenya: a multihospital retrospective cohort study

OBJECTIVES

To characterise adoption and explore specific clinical and patient factors that might influence pulse oximetry and oxygen use in low-income and middle-income countries (LMICs) over time; to highlight useful considerations for entities working on programmes to improve access to pulse oximetry and oxygen.

DESIGN

A multihospital retrospective cohort study.

SETTINGS

All admissions (n=132 737) to paediatric wards of 18 purposely selected public hospitals in Kenya that joined a Clinical Information Network (CIN) between March 2014 and December 2020.

OUTCOMES

Pulse oximetry use and oxygen prescription on admission; we performed growth-curve modelling to investigate the association of patient factors with study outcomes over time while adjusting for hospital factors.

RESULTS

Overall, pulse oximetry was used in 48.8% (64 722/132 737) of all admission cases. Use rose on average with each month of participation in the CIN (OR: 1.11, 95% CI 1.05 to 1.18) but patterns of adoption were highly variable across hospitals suggesting important factors at hospital level influence use of pulse oximetry. Of those with pulse oximetry measurement, 7% (4510/64 722) had hypoxaemia (SpO2 <90%). Across the same period, 8.6% (11 428/132 737) had oxygen prescribed but in 87%, pulse oximetry was either not done or the hypoxaemia threshold (SpO2 <90%) was not met. Lower chest-wall indrawing and other respiratory symptoms were associated with pulse oximetry use at admission and were also associated with oxygen prescription in the absence of pulse oximetry or hypoxaemia.

CONCLUSION

The adoption of pulse oximetry recommended in international guidelines for assessing children with severe illness has been slow and erratic, reflecting system and organisational weaknesses. Most oxygen orders at admission seem driven by clinical and situational factors other than the presence of hypoxaemia. Programmes aiming to implement pulse oximetry and oxygen systems will likely need a long-term vision to promote adoption, guideline development and adherence and continuously examine impact.

The Simple One-Step (SOS) Stool Processing Method for Use with the Xpert MTB/RIF Assay for a Child-Friendly Diagnosis of Tuberculosis Closer to the Point of Care

Young children cannot easily produce sputum for diagnosis of pulmonary tuberculosis (TB). Alternatively, Mycobacterium tuberculosis complex bacilli can be detected in stool by using the Xpert MTB/RIF (Ultra) assay (Xpert). Published stool processing methods contain somewhat complex procedures and require additional supplies. The aim of this study was to develop a simple one-step (SOS) stool processing method based on gravity sedimentation only, similar to Xpert testing of sputum samples, for the detection of M. tuberculosis in stool samples. We first assessed whether the SOS stool method could provide valid Xpert results without the need for bead-beating, dilution, and filtration steps. We concluded that this was the case, and we then validated the SOS stool method by testing spiked stool samples. By using the SOS stool method, 27 of the 29 spiked samples gave valid Xpert results, and M. tuberculosis was recovered from all 27 samples. The proof of principle of the SOS stool method was demonstrated in routine settings in Addis Ababa, Ethiopia. Nine of 123 children with presumptive TB had M. tuberculosis-positive results for nasogastric aspiration (NGA) samples, and 7 (77.8%) of those children also had M. tuberculosis-positive Xpert results for stool samples. Additionally, M. tuberculosis was detected in the stool samples but not the NGA samples from 2 children. The SOS stool processing method makes use of the standard Xpert assay kit, without the need for additional supplies or equipment. The method can potentially be rolled out to any Xpert site, bringing a bacteriologically confirmed diagnosis of TB in children closer to the point of care.

Employing learning health system principles to advance research on severe neonatal and paediatric illness in Kenya

We have worked to develop a Clinical Information Network (CIN) in Kenya as an early form of learning health systems (LHS) focused on paediatric and neonatal care that now spans 22 hospitals. CIN's aim was to examine important outcomes of hospitalisation at scale, identify and ultimately solve practical problems of service delivery, drive improvements in quality and test interventions. By including multiple routine settings in research, we aimed to promote generalisability of findings and demonstrate potential efficiencies derived from LHS. We illustrate the nature and range of research CIN has supported over the past 7 years as a form of LHS. Clinically, this has largely focused on common, serious paediatric illnesses such as pneumonia, malaria and diarrhoea with dehydration with recent extensions to neonatal illnesses. CIN also enables examination of the quality of care, for example that provided to children with severe malnutrition and the challenges encountered in routine settings in adopting simple technologies (pulse oximetry) and more advanced diagnostics (eg, Xpert MTB/RIF). Although regular feedback to hospitals has been associated with some improvements in quality data continue to highlight system challenges that undermine provision of basic, quality care (eg, poor access to blood glucose testing and routine microbiology). These challenges include those associated with increased mortality risk (eg, delays in blood transfusion). Using the same data the CIN platform has enabled conduct of randomised trials and supports malaria vaccine and most recently COVID-19 surveillance. Employing LHS principles has meant engaging front-line workers, clinical managers and national stakeholders throughout. Our experience suggests LHS can be developed in low and middle-income countries that efficiently enable contextually appropriate research and contribute to strengthening of health services and research systems.

Improving case detection of tuberculosis in hospitalised Kenyan children-employing the behaviour change wheel to aid intervention design and implementation

BACKGROUND

The true burden of tuberculosis in children remains unknown, but approximately 65% go undetected each year. Guidelines for tuberculosis clinical decision-making are in place in Kenya, and the National Tuberculosis programme conducts several trainings on them yearly. By 2018, there were 183 GeneXpert® machines in Kenyan public hospitals. Despite these efforts, diagnostic tests are underused and there is observed under detection of tuberculosis in children. We describe the process of designing a contextually appropriate, theory-informed intervention to improve case detection of TB in children and implementation guided by the Behaviour Change Wheel.

METHODS

We used an iterative process, going back and forth from quantitative and qualitative empiric data to reviewing literature, and applying the Behaviour Change Wheel guide. The key questions reflected on included (i) what is the problem we are trying to solve; (ii) what behaviours are we trying to change and in what way; (iii) what will it take to bring about desired change; (iv) what types of interventions are likely to bring about desired change; (v) what should be the specific intervention content and how should this be implemented?

RESULTS

The following behaviour change intervention functions were identified as follows: (i) training: imparting practical skills; (ii) modelling: providing an example for people to aspire/imitate; (iii) persuasion: using communication to induce positive or negative feelings or stimulate action; (iv) environmental restructuring: changing the physical or social context; and (v) education: increasing knowledge or understanding. The process resulted in a multi-faceted intervention package composed of redesigning of child tuberculosis training; careful selection of champions; use of audit and feedback linked to group problem solving; and workflow restructuring with role specification.

CONCLUSION

The intervention components were selected for their effectiveness (from literature), affordability, acceptability, and practicability and designed so that TB programme officers and hospital managers can be supported to implement them with relative ease, alongside their daily duties. This work contributes to the field of implementation science by utilising clear definitions and descriptions of underlying mechanisms of interventions that will guide others to do likewise in their settings for similar problems.

Perspectives and practices of health workers around diagnosis of paediatric tuberculosis in hospitals in a resource-poor setting - modern diagnostics meet age-old challenges

BACKGROUND

Detection of tuberculosis (TB) in children in Kenya is sub-optimal. Xpert MTB/RIF® assay (Xpert®) has the potential to improve speed of TB diagnosis due to its sensitivity and fast turnaround for results. Significant effort and resources have been put into making the machines widely available in Kenya, but use remains low, especially in children. We set out to explore the reasons for the under-detection of TB and underuse of Xpert® in children, identifying challenges that may be relevant to other newer diagnostics in similar settings.

METHODS

This was an exploratory qualitative study with an embedded case study approach. Data collection involved semi-structured interviews; small-group discussions; key informant interviews; observations of TB trainings, sensitisation meetings, policy meetings, hospital practices; desk review of guidelines, job aides and policy documents. The Capability, Opportunity and Motivation (COM-B) framework was used to interpret emerging themes.

RESULTS

At individual level, knowledge, skill, competence and experience, as well as beliefs and fears impacted on capability (physical & psychological) as well as motivation (reflective) to diagnose TB in children and use diagnostic tests. Hospital level influencers included hospital norms, processes, patient flows and resources which affected how individual health workers attempted to diagnose TB in children by impacting on their capability (physical & psychological), motivation (reflective & automatic) and opportunity (physical & social). At the wider system level, community practices and beliefs, and implementation of TB programme directives impacted some of the decisions that health workers made through capability (psychological), motivation (reflective & automatic) and opportunity (physical).

CONCLUSION

We used comprehensive approaches to identify influencers of TB case detection and use of TB diagnostic tests in children in Kenya. These results are being used to design a contextually-appropriate intervention to improve TB diagnosis, which may be relevant to similar low-resource, high TB burden countries and can be feasibly implemented by the National TB programme.

Handling missing data in modelling quality of clinician-prescribed routine care: Sensitivity analysis of departure from missing at random assumption

Missing information is a major drawback in analyzing data collected in many routine health care settings. Multiple imputation assuming a missing at random mechanism is a popular method to handle missing data. The missing at random assumption cannot be confirmed from the observed data alone, hence the need for sensitivity analysis to assess robustness of inference. However, sensitivity analysis is rarely conducted and reported in practice. We analyzed routine paediatric data collected during a cluster randomized trial conducted in Kenyan hospitals. We imputed missing patient and clinician-level variables assuming the missing at random mechanism. We also imputed missing clinician-level variables assuming a missing not at random mechanism. We incorporated opinions from 15 clinical experts in the form of prior distributions and shift parameters in the delta adjustment method. An interaction between trial intervention arm and follow-up time, hospital, clinician and patient-level factors were included in a proportional odds random-effects analysis model. We performed these analyses using R functions derived from the jomo package. Parameter estimates from multiple imputation under the missing at random mechanism were similar to multiple imputation estimates assuming the missing not at random mechanism. Our inferences were insensitive to departures from the missing at random assumption using either the prior distributions or shift parameters sensitivity analysis approach.