Falls in the elderly are common and can have adverse physical and psychological consequences.
The etiology of falls is often multi-factorial. Risk factors include age over 75 years, three or more medications, recent changes in medication or mental state, lower limb problems, visual impairment, cardiovascular illness, dementia, and depression (Lim et al., Reference Lim, Ng, Ng and Ng2001).
We describe a case of psychotically-driven falls, an unusual and uncommon cause of falls in the elderly not previously described in the literature.
The patient was an 85 year old woman with a diagnosis of chronic schizoaffective disorder. She presented with persecutory delusions that nursing home staff were poisoning her food, erotomanic delusions about her psychogeriatrician, and command auditory hallucinations ordering her to fall.
The hallucinations consisted of unfamiliar and threatening male voices, telling her that “if you do not fall now, you will fall worse later.” At other times, the voice would be that of her geriatric psychiatrist telling her that she needed to prove her love to him by falling.
She complied with these commands and would fall on average twice weekly. As a result she sustained a large periorbital hematoma and multiple contusions to her upper and lower limbs.
Witnesses to the falls reported that she would be sitting in a chair and would shuffle her bottom forwards until she slid off the chair onto the floor. Alternatively, she would slide off her bed onto the floor. Significantly, the falls never occurred whilst walking and she never lost consciousness.
Her psychotropic medications were olanzapine 12.5 mg daily and depot pipothiazine palmitate (a phenothiazine antipsychotic) 75 mg three weekly. The olanzapine had been reduced to 5 mg daily three weeks prior to admission by her community geriatric psychiatrist because of extrapyramidal side effects. Her mental state subsequently deteriorated, with development of auditory hallucinations, necessitating her admission to hospital. Other relevant medications were diltiazem, cilazapril, budesonide/eformoterol, and salbutamol.
The patient was a resident of a nursing home and shared a room with another resident. She required assistance from staff for all her activities of daily living except for eating and drinking, in which she was independent. She used a walking frame to mobilize. She was well supported by a daughter, who was her enduring power of attorney, and by a case worker, who visited regularly.
She had no significant cognitive deficits and scored 29/30 on Mini-Mental State Examination, losing one point on delayed recall of three objects.
Although she had other risk factors for falls, such as age, polypharmacy, psychotropic drugs, antihypertensive medication, and reduced mobility, examination revealed nothing to provide a physical explanation for her falls. Furthermore, her falls ceased as her auditory hallucinations improved with pharmacotherapy. We concluded, therefore, that psychosis was the cause.
Command hallucinations are common in patients with auditory hallucinations. In an Australian study, two thirds of a sample of 193 patients with auditory hallucinations experienced command hallucinations (MacKinnon et al., Reference MacKinnon, Copolov and Trauer2004). Elderly patients with auditory hallucinations are just as likely to have command hallucinations and are more likely to obey them, especially if the voices are considered benevolent (Cohen et al., Reference Cohen, Izediuno, Yadack, Ghosh and Garrett2014).
The commands are often violent: in one sample, 44% reported commands of injury to self or others (Zisook et al., Reference Zisook, Byrd, Kuck and Jeste1995). Compliers are more likely to know the voice, to be emotionally involved during the hallucination, and to consider the voices real. These three characteristics were present in this case.
To our knowledge, this is the first report of falls in the elderly caused by command hallucinations. Although uncommon, psychosis should be included as a potential cause of falls in the psycho-geriatric population.
Conflict of interest
None.
Description of authors’ roles
Dr Payman suggested the idea of writing the case vignette to Dr Glauser, who collected the information and assisted Dr Payman with the writing of the vignette.
Acknowledgments
Dr Kate Mullin and Dr Kevin Ong for comments on the manuscript. We are also grateful to the patient, from whom we obtained written consent for publication of her case.