[Skip to Content]
[Skip to Content Landing]
September 10, 1997

Detention of Persistently Nonadherent Patients With Tuberculosis

Author Affiliations

From the Program in Medical Ethics and the Division of General Internal Medicine (Mr Oscherwitz, Drs Tulsky and Lo, and Ms Alpers), the Center for AIDS Prevention Studies (Ms Alpers and Dr Lo), University of California, San Francisco, and the California Department of Health Services, Berkeley (Messrs Roger and Sciortino and Dr Royce).

JAMA. 1997;278(10):843-846. doi:10.1001/jama.1997.03550100069041

Context.  —Patient with tuberculosis (TB) who are persistently nonadherent to treatment present a public health risk. In 1993, California created a new civil detention process and allowed detention of noninfectious but persistently nonadherent patients.

Objectives.  —To determine (1) which patients TB controllers attempt to detain, (2) how often and where patients are detained, and (3) how many of these patients complete TB treatment.

Design.  —Case series with cross-sectional comparison to other adult TB patients in the study counties.

Setting.  —Twelve California counties with the largest number of new TB cases reported in 1994.

Subjects.  —All patients whom TB controllers sought to detain during 1994 and 1995 because of persistent nonadherence to treatment.

Data Sources.  —Public health records, interviews with county TB officials, and Reports of Verified Cases of Tuberculosis to the California Tuberculosis Control Branch.

Results.  —Tuberculosis controllers sought the civil detention or arrest of 67 patients during the study period (1.3% of adult TB patients with the same disease sites). Forty-six percent of these patients were homeless, 81% had drug or alcohol abuse, and 28% had mental illness. Tuberculosis controllers sought civil detention of 15 patients. Fourteen patients were detained (median length of detention, 14.5 days). Tuberculosis controllers sought to arrest 62 patients during the study period. Fifty-three patients were arrested (median time in jail, 83 days). In 10 cases, both civil and criminal detention were attempted. We analyzed completion of therapy after excluding patients who were not detained or who died or moved. Overall, 41 (84%) of the remaining 49 detained patients completed therapy. Of the patients who completed therapy, only 17 were detained until treatment was completed. Compared with other TB patients in these counties, detained patients had 4 times the proportion lost to follow-up and half the proportion completing therapy within 12 months.

Conclusion.  —Further improvements in the care of persistently nonadherent patients may require more psychosocial services, appropriate facilities for civil detention, and detaining patients long enough to assure completion of treatment.