A longitudinal and case-control study of dropout among drug users in methadone maintenance treatment in Haiphong, Vietnam.

Background: Vietnam began providing methadone maintenance therapy (MMT) in 2008; as of June 2016, 44,479 persons who inject drugs (PWID) were in treatment in 57 provinces. However, 10-23% of patients were estimated to have dropped out of treatment during the first 2 years. We evaluated dropout and factors associated with quitting treatment.

Methods: We followed clients ≥ 18 years old enrolled in five MMT clinics in Haiphong for 3 years. Persons who missed a consecutive month of methadone treatment were considered to have dropped out and were not allowed to return; those who missed greater than five consecutive doses were considered to be non-compliant but were allowed to restart treatment at their initial dose. Clients who dropped out or who were non-compliant during their third year of MMT (cases) were traced and matched with two clients who remained in treatment (controls) by gender, age, and length of time in the program. Cases and controls were interviewed. Additional data on levels of yearly retention were abstracted from clinic records.

Results: Among the 1055 patients initially enrolled in MMT, dropout and non-compliance combined was 13.6% during the first year, 16.5% during the second year, and 22.3% during the third year. By 36 months, 33.3% of clients had dropped out, of whom 10.6% had died and 24% had been arrested. We traced and interviewed 81 clients who dropped out or who were non-compliant during year 3 as well as 161 controls. The primary reasons for dropping out included claiming no dependence on heroin (22.2%), conflict with work (21.0%), health problems (16.0%), and inability to afford the methadone co-payment of approximately 0.5 USD/day (14.8%). Independent factors associated with non-compliance included continuing to use heroin (aOR = 12.4, 95% CI 4.2-36.8) and missing greater than three doses during the previous 3 months (aOR = 18.5, 95% CI 7.4-47.1); receiving a daily dose of > 120 mg of methadone was associated with a lower odds ratio of dropping out (aOR = 0.3, 95% CI 0.1-0.9).

Conclusion: By 3 years, one third of all patients in treatment had permanently dropped out. Ensuring that methadone dosing is adequate and reducing or eliminating the co-payment fee for those who cannot afford it could improve retention.