Skip to main content

Table 2 Multivariable logistic regression of the association between clinician caseload and treatment outcome

From: Towards better guidance on caseload thresholds to promote positive tuberculosis treatment outcomes: a cohort study

Main exposure/confounders

Multivariable regressiona

OR (95 % CI)

Clinician caseloadb

 

10+

P = 0.002

 

<10

1.14 (1.05–1.25)

Notification date

 

Post-toolkit

P <0.001

 

Pre-toolkit

1.33 (1.23–1.45)

Location

 

Outside London

P <0.001

 

Inside London

0.82 (0.74–0.91)

Sex

 

Male

P <0.001

 

Female

0.85 (0.80–0.91)

Age, years

 

<20

0.72 (0.63–0.82)

 

20–39

P <0.001

 

40–64

0.88 (0.81–0.95)

 

65+

1.30 (1.18–1.44)

Ethnic group

 

White

P <0.001

 

Black African

0.79 (0.71–0.88)

 

Black other

0.88 (0.72–1.07)

 

Indian subcontinent

0.86 (0.79–0.94)

 

Other

0.86 (0.76–0.97)

Previous diagnosis

 

No

P <0.001

 

Yes

1.26 (1.12–1.42)

Social risk factorsc

 

No or unknown

P <0.001

 

One or more previous

1.44 (1.16–1.77)

 

One or more current

2.56 (2.14–3.07)

Shared management

 

No

P = 0.08

 

Yes

1.37 (0.97–1.94)

  1. Multivariable random effects logistic regression of the association between clinician caseload and treatment outcome, England, 2006–2012. Model adjusted for clustering by clinician and the confounders in the Table; 34,707 records in this model
  2. aOdds of an unfavourable versus a good or neutral treatment outcome
  3. bMean caseload per clinician over the preceding 3 years
  4. cSocial risk factors a composite variable of homelessness, imprisonment, drug misuse, and alcohol abuse; current risks override previous risks
  5. CI Confidence interval, OR Cluster-specific odds ratio, P, P value
Back to article page