Table 5

 Unadjusted and adjusted hazard ratios for chlamydia reinfection among 16–24-year-old women by setting of recruitment and population characteristics, England, March 2002–August 2003

CharacteristicRecruited from general practiceRecruited from family planningRecruited from GUMTotal stratified by healthcare setting of recruitment
Unadjusted hazard ratios (95% CI)Unadjusted hazard ratios (95% CI)Unadjusted hazard ratios (95% CI)Adjusted hazard ratios* (95% CI)
GUM, genitourinary medicine; ref, reference group.
*Adjusted hazard ratios come from a multivariable Cox regression stratified by healthcare setting of recruitment.
†Variable selection in adjusted stratified Cox regression model was influenced by unadjusted hazard ratios (p<0.10) and epidemiological factors previously found to be associated with reinfection.
‡Last adjusted hazard ratio before the variable was dropped from multivariable modelling.
§All symptoms dropped from the final model as they are indicators of infection and not risk factors for acquisition.
¶The rate of new sexual partner change for the time period immediately preceding infection.
**Rate of new sexual partner change as continuous.
††Rate of new sexual partner change as a continuous variable was dropped from model, as the categorical variable provided more explanatory power in the final adjusted model.
‡‡Rate uses known partners as denominator.
§§Rate uses reported partners as denominator.
¶¶Effective partner treatment rate used in adjusted stratified model as standard partner follow-up activities are based on what the patient reports to the nurse.
***Reference group in multivariable model changed to use the lowest-risk group (used condoms, not all partners treated) as reference.
NB: No evidence found for non-proportional hazards in the final multivariable model (p = 0.23). No evidence of a difference between health care settings of recruitment (in a model with health care setting of recruitment as a covariate rather than as a stratification variable; p = 0.18).
TotalRef0.8 (0.5 to 1.4)p = 0.3960.7 (0.4 to 1.3)p = 0.285
Reason for attendance (baseline)
    Attended for symptoms0.7 (0.1 to 5.3)p = 0.7160.7 (0.2 to 2.5)p = 0.6151.8 (0.7 to 4.7)p = 0.211
    Other reason, but reported symptoms2.0 (0.9 to 4.8)p = 0.1080.8 (0.4 to 1.8)p = 0.5981.2 (0.5 to 3.0)p = 0.683
    Asymptomaticref.ref.ref.
Age group (years)†
    16–172.8 (0.9 to 8.1)p = 0.0641.8 (0.6 to 5.0)p = 0.2812.5 (0.8 to 8.1)p = 0.1351.9 (0.9 to 4.1)p = 0.116
    18–200.8 (0.3 to 2.2)p = 0.7281.2 (0.4 to 3.5)p = 0.6791.7 (0.6 to 4.7)p = 0.2981.1 (0.5 to 2.3)p = 0.772
    21–24RefRefRefRef
Ethnic group†
    WhiteRefRefRefRef
    Non-whiteNo eventsNo eventsNo eventsNo events
    Unknown0.7 (0.2 to 1.8)p = 0.4050.6 (0.1 to 4.6)p = 0.6440.4 (0.1 to 1.4)p = 0.175Not in model
Ever had chlamydia (baseline)
    Yes0.6 (0.1 to 4.7)p = 0.648No events1.8 (0.7 to 4.9)p = 0.236
    NoRefRefRef
Condom use at last sexual intercourse†
    Yes0.8 (0.2 to 3.3)p = 0.7101.3 (0.5 to 3.2)p = 0.5520.9 (0.3 to 3.1)p = 0.8560.7 (0.3 to 1.6)p = 0.487‡
    NoRefRefRefref.
Symptoms (yes, past 3 or 6 months)§No is refNo is refNo is ref
    Cystitis (urinary tract infection)2.5 (0.9 to 6.6)p = 0.0651.1 (0.4 to 3.3)p = 0.8420.7 (0.2 to 2.9)p = 0.598
    Frequent urination1.5 (0.4 to 5.1)p = 0.5442.1 (0.7 to 6.4)p = 0.1751.1 (0.4 to 3.2)p = 0.896
    Abdominal pain1.6 (0.5 to 4.9)p = 0.3941.0 (0.3 to 3.0)p = 0.9874.1 (1.7 to 10.0)p = 0.002
    Bleeding between periods1.3 (0.4 to 4.6)p = 0.6412.4 (0.8 to 7.1)p = 0.1122.0 (0.7 to 6.1)p = 0.214
    Bleeding after sexual intercourse2.3 (0.5 to 10.0)p = 0.2751.7 (0.2 to 12.4)p = 0.6214.1 (1.3 to 12.5)p = 0.013
    Pain during sexual intercourse0.6 (0.1 to 4.9)p = 0.6750.8 (0.2 to 3.3)p = 0.7383.0 (1.1 to 7.7)p = 0.025
    Discharge1.4 (0.5 to 4.2)p = 0.5511.5 (0.5 to 4.0)p = 0.4452.1 (0.9 to 5.2)p = 0.097
    None0.8 (0.2 to 2.3)p = 0.7030.6 (0.2 to 1.5)p = 0.2930.4 (0.1 to 1.1)p = 0.079
Recent rate of partner change per 6 months†, ¶4.1 (0.7 to 24.4)p = 0.123**3.5 (1.4 to 8.9)p = 0.007**1.7 (0.5 to 5.7)p = 0.427**††
    No male sexual partner changeRefRefRefRef
    1 new male sexual partnersNo events5.1 (1.4 to 18.5)p = 0.0120.6 (0.1 to 4.5)p = 0.6001.2 (0.5 to 3.1)p = 0.702‡
    2 new male sexual partners3.8 (1.4 to 10.2)p = 0.0134.2 (1.2 to 14.7)p = 0.0242.9 (1.2 to 7.3)p = 0.0202.9 (1.5 to 5.6)p = 0.002
    3 new male sexual partners5.6 (0.6 to 49.5)p = 0.12013.9 (3.0 to 65.1)p = 0.0012.2 (0.3 to 17.1)p = 0.4684.1 (1.4 to 11.8)p = 0.010
    ⩾4 new male sexual partners1.1 (0.1 to 9.2)p = 0.9185.2 (1.1 to 23.2)p = 0.0320.9 (0.1 to 7.1)p = 0.9221.4 (0.5 to 3.8)p = 0.532‡
Patient management for initial positive test
Patient treatment of initial positive test
    AzithromycinRefRefRef
    Other regimen0.3 (0.03 to 1.9)p = 0.1770.9 (0.3 to 2.7)p = 0.9101.1 (0.5 to 2.4)p = 0.886
Partner management details for initial positive test
Partner treatment rate‡‡
    100% of known partners treatedRefRefRef
    <100% of known partners treated1.8 (0.8 to 4.1)p = 0.1762.6 (1.1 to 6.0)p = 0.0301.5 (0.7 to 3.3)p = 0.302
Effective partner treatment rate†,§§¶¶
    100% of reported partners treatedRefRefRefRef
    <100% of reported partners treated2.4 (1.0 to 5.6)p = 0.0472.3 (1.1 to 4.9)p = 0.0371.8 (0.8 to 3.9)p = 0.1311.8 (1.0 to 3.3)p = 0.034‡
Effective partner treatment rate§§ and condom use at last sex†
    Used condom at last sexual intercourse
    All reported partners treatedRefRefRef4.7 (0.9 to 23.3) p = 0.058
    Not all reported partners treatedNo events0.4 (0.1 to 2.1)p = 0.286No eventsRef***
Did not use a condom at last sexual intercourse
    All reported partners treated0.6 (0.1 to 3.0)p = 0.5730.3 (0.1 to 0.9)p = 0.0410.3 (0.1 to 1.1)p = 0.0712.0 (0.4 to 9.5)p = 0.377
    Not all reported partners treated1.5 (0.3 to 7.3)p = 0.6071.0 (0.3 to 2.9)p = 0.9881.0 (0.3 to 3.6)p = 0.9975.8 (1.3 to 25.9)p = 0.021