Current issues of ACP Journal Club are published in Annals of Internal Medicine


Home visits for elderly persons after hospital discharge did not reduce mortality or hospital readmissions but reduced the use of institutional care

ACP J Club. 1993 May-June;118:79. doi:10.7326/ACPJC-1993-118-3-079

Source Citation

Hansen FR, Spedtsberg K, Schroll M. Geriatric follow-up by home visits after discharge from hospital: a randomized controlled trial. Age Ageing. 1992 Nov;21:445-50.



To evaluate the effect of home visits by district nurses and general practitioners on mortality and institutionalization for elderly persons discharged from the hospital.


1-year controlled trial with treatment assignment based on birth date.


A municipality in Denmark and its district hospital.


All persons ≥ 75 years old who were hospitalized from their own homes. 389 of 404 eligible patients started the study, and 344 (24% aged ≥ 80 y, 233 men) completed it. 45 persons were rehospitalized within 14 days and were not included in the final study group.


181 participants born on an even date were allocated to usual social and medical care at discharge. 163 participants born on an odd date were to receive a home visit from a nurse on the day after discharge and a visit from their general practitioner 2 weeks later. The nurse and physician evaluated follow-up care, adjusted medications, and arranged for further care if needed. All care for both groups used existing local resources.

Main outcome measures

After 1 year, participants were contacted to ascertain mortality, nursing home use, and readmissions to the hospital.

Main results

Analysis was by intention to treat. In the home-visit group, 108 participants had both visits, 43 had the nurse visit only, and 12 had the physician visit only. During the nurse visits 70 participants (46%) had new or unforeseen problems, and 51 (34%) needed additional care such as a home health aide or meal delivery. Physicians adjusted medications in 48% of the participants and found major social problems in 13%. The groups did not differ for mortality (Table) or for readmissions to the hospital (both groups had a 1-year readmission rate of 46%). Compared with participants in the control group, participants in the visits group had fewer admissions to nursing homes (P < 0.05) during the first year of discharge but not at 1 year of discharge (P = 0.09)*.


Follow-up visits from a district nurse and a general practitioner to persons aged ≥ 75 years and recently discharged from the hospital reduced the use of institutional care but had no effect on mortality or hospital readmission.

Sources of funding: Danish Ministry of Health and the Medical Research Foundation for region 3.

For article reprint: Dr. F.R. Hansen, The County Hospital of Roskilde, Frølichsvej 29, DK-2920 Charlottenlund, Denmark. FAX 45-46-32-1055.

*P value calculated from data in article.

Table. Home visits vs usual care after discharge from hospital in elderly persons†

Outcomes Home visits Usual care RRR (95% CI) NNT (CI)
Mortality at 1 year 20% 24% 17% (-23 to 45) Not significant
Admissions to nursing home at 1 year 10% 16% 39% (-7 to 65) Not signficant
Admission to nursing home during 1 year 6% 13% 54% (8 to 77) 15 (8 to 117)

†Abbreviations defined in Glossary; RRR, NNT, and CI calculated from data in articles.


In the United States, less than 2% of encounters between patients and primary care physicians occur at home. Physicians are, however, becoming increasingly involved with home care, often spending several hours per week doing telephone consultation or completing paperwork (1). Despite the dramatic increase in the volume and technologic sophistication of home care, its effectiveness has had little study.

This study from Denmark, despite some methodologic flaws (lack of blinding, high rate of exclusion because of early rehospitalization, a surprisingly low percentage of older women), offers several important insights. First, to be effective, home care need not be resource- or technology-intensive. This study used existing community resources and a short-term case-management model to delay the need for institutionalization. Second, targeting interventions to those at highest risk (in this case, the most elderly patients) is likely to lead to the greatest efficacy. This is confirmed by a recent study of Department of Veterans Affairs-based home care (2). Third, home assessments can be uniquely sensitive diagnostic tools. In this study, many unforeseen problems were identified by visiting nurses and physicians despite a thorough evaluation during a recent hospitalization.

It remains to be seen whether these encouraging findings will translate into public policy that lowers current barriers to more meaningful participation by primary care physicians in home care.

Joseph Francis, MD, MPH
Department of Veterans Affairs Medical Center Memphis, Tennessee, USA