Logo of geronaLink to Publisher's site
J Gerontol A Biol Sci Med Sci. Oct 2011; 66A(10): 1124–1130.
Published online Jul 15, 2011. doi:  10.1093/gerona/glr113
PMCID: PMC3202899

Antidepressant Prescriptions: An Acute Window for Falls in the Nursing Home

Abstract

Background.

Although many studies have implicated antidepressants as a risk factor for falls, it is not clear if risk accrues with duration of use or if there are acute risks associated with initiation of the prescription. We conducted a case-crossover study of nursing home residents with a fall to determine the effect of an antidepressant change (defined as the new prescription of an antidepressant or increasing the dose of a previously used antidepressant) on fall risk.

Methods.

Among 1,181 nursing home fallers, we compared the frequency of antidepressant changes during the hazard period (1–7 days before the fall) with the frequency of antidepressant changes during the control period (8–14 days before the fall). Odds ratios were estimated using conditional logistic regression models. Results were estimated for non-selective serotonin reuptake inhibitors (SSRI) and SSRI prescriptions, separately.

Results.

Mean age was 88 years, and 71% were females. Seventy participants experienced an antidepressant change during the hazard and/or control periods. The maximum effect of falling occurred within 2 days of a non-SSRI change (odds ratio: 4.7, 95% confidence interval, 1.3–16.2). The effect on falling was no longer significant at 5 days (odds ratio: 1.9, 95% confidence interval, 0.9–4.0). No association was found between SSRI changes and falls.

Conclusions.

Nursing home residents are at high risk of falls during the days following a new prescription or increased dose of a non-SSRI antidepressant. Increased surveillance should occur, particularly during the first 48 hours, in an effort to decrease falls.

Keywords: Antidepressant, Fall, Nursing home

FALLS occur commonly in the nursing home at an average rate of 1.5 falls per resident annually (1). Among nursing home residents, falls may result in fractures, disability, and even death (24). Medication use is common and a potentially modifiable risk factor for falls (3,5). Both tricyclic antidepressants (TCA) and selective serotonin reuptake inhibitors (SSRI) have been associated with a 1.4- to 6-fold increased risk of falls (6,7). It is less clear whether newer antidepressants, such as serotonin–norepinephrine reuptake inhibitors, might also be associated with an increased risk of falls (6,8,9). Regardless, antidepressants are effective in remediating depressive symptoms, and few clinicians chose to withhold these drugs based on risk of falls. Identifying chronic antidepressant users is unlikely to be helpful in reducing falls in the nursing home setting, given the high prevalence and long anticipated duration of use.

In contrast, users of antidepressants may be at a particularly high risk of falls in the days following a new prescription or increased dose of a previously used antidepressant. Identifying windows of time associated with an acute increased risk of falls is important because this is likely to lead to the development of successful fall intervention programs. Therefore, we conducted a case-crossover study of nursing home residents with a fall to examine the effect of a new prescription or increasing the dose of a previously used antidepressant on the risk of falls. Because the pharmacologic mechanisms and side effects of antidepressants vary, we examined the effect of a new prescription or increased dose of all antidepressant classes on risk of falls together, as well as SSRI and non-SSRI antidepressant classes separately.

METHODS

Study Design

The case-crossover design was specifically developed to assess the effects of a transient exposure on an acute event (10). In the case-crossover design, the frequency of exposure during a selected period of time preceding the event (hazard period) is compared with the frequency of usual exposure (control period) in the same individual. Thus, all cases serve as their own controls. This lessens control selection bias; removes time invariant confounding, such as comorbidities; and minimizes the need to use multivariable adjustment models to account for differences between persons.

Although this study design is typically not appropriate to study medications that are used on a regular basis, it may be used to assess associations related to a change in a regularly used medication (ie, moving from no medication use to medication use or from low medication use to high medication use). The case-crossover study design has been used to assess both the effects of intermittent medication use (11,12) and medication changes (13,14) on the occurrence of acute events.

We hypothesized that antidepressant changes might cause sedation or other adverse effects prior to reaching a steady state or therapeutic effect. Thus, we selected the hazard period a priori to be between 1 and 7 days before the fall. We selected a control period exactly 7 days before the hazard period in the same individual to minimize differences in underlying risk of falls across the hazard and control periods. We then compared the occurrence of antidepressant changes during the hazard period (1–7 days before the fall) with their occurrence during the control period (8–14 days before the fall).

Participants

Participants were 1,181 long-term care residents of two large long-term care facilities (Hebrew SeniorLife) in Boston, MA, who experienced a fall between January 2005 and May 2010. In order to be certain that we could fully ascertain exposure to medication prescriptions during the hazard and control periods, we only included the first fall a participant experienced greater than or equal to 15 days from admission to the long-term care setting. Residents less than age 60 years were excluded (n = 10). This study was approved by the Institutional Review Board of Hebrew SeniorLife.

Falls

Falls were ascertained prospectively using the facility’s federally mandated, computerized incident reports. Nursing staff are trained annually in incident reporting, with a fall defined as accidently coming to rest on the ground or lower surface. Because the majority of falls in a nursing home are unwitnessed, it is possible that a fall could be unreported if the resident was able to rise without assistance and did not report the injury.

Antidepressant Changes

All changes in antidepressants, defined as either a new prescription or dose increase of a previously used antidepressant, were ascertained 1–14 days before the fall using the facility’s computerized order entry system (Meditech). A “new prescription” was defined as an order for any antidepressant that was not prescribed 15 days before the fall. Switching prescriptions within a class of medication (ie, from trazodone to mirtazapine) was considered a new prescription as well. A “dose increase” was defined as either an increase in the dose or frequency of an antidepressant that was prescribed 15 days before the fall. Decreasing the dose of an antidepressant was not considered an event of interest. Medications were categorized as SSRI or non-SSRI (“miscellaneous antidepressants”) based on the American Hospital Formulary Services (15).

Other Characteristics of Interest

Given our study design, which compared each participant with himself or herself 1 week prior, it was not necessary to use multivariable regression to adjust for traditional risk factors for falls, such as age or comorbidities. However, in order to describe our population, we collected information on additional characteristics of interest.

Medication count, defined as the number of medications a resident was taking 15 days prior to the fall, was ascertained using the computer order entry system. We defined a medication as any over-the-counter or prescription medication that was ordered scheduled or as needed. We excluded topicals, vitamins, minerals, and over-the-counter bowel medications. Residents prescribed more than 8 baseline medications were classified as using a large number of medications.

Information on all additional characteristics of interest was obtained using the Minimum Data Set assessment closest to and preceding the fall. The Minimum Data Set is a federally mandated needs assessment that is performed on all nursing home residents at the time of admission and quarterly thereafter (16). Cognitive status was described using the validated Cognitive Performance Scale (0–6) (17). Scores were categorized as intact or mild impairment (Scores 0–1), moderate impairment (Scores 2–4), or severe impairment (Scores 5–6). Functional status was described using the modified Katz ADL short scale (18) and categorized as intact or mild impairment (Scores 0–1), moderate impairment (Scores 2–3), or severe impairment (Scores 4–7). Depression was measured using a validated Minimum Data Set Depression Rating Scale, which is sensitive to detecting depression among cognitively impaired persons (19). This scale queries whether seven depressive symptoms were never (0), occasionally (1), or always (2) present in nursing home residents within the past 30 days. Scores from the seven items are summed, and we categorized depressive symptoms as none (0), mild (Scores 1–2), or moderate to severe (Score ≥ 3). History of recent fall was ascertained by nursing report of falls in the past 90 days. Active comorbidities, including anemia, arthritis, congestive heart failure, diabetes, emphysema, hypertension, and stroke, were described.

Statistical Analysis

We estimated odds ratios of falling following the new prescription or increased dose of an antidepressant by comparing the occurrence of antidepressant changes during the hazard period with its occurrence during the control period using conditional logistic regression models. Because the effect of an antidepressant change may differ by baseline characteristics, we stratified our analyses based on the median number of baseline medications used and also by recent history of falls. If estimates appeared different, a formal interaction term was added to the model.

Since the true effect period during which antidepressant use affects falls is unknown, we varied the length of the hazard and control periods. This method has been recommended in situations where the biological effects of the exposure are unknown (10,20). Given our hypothesis, we chose hazard periods from 2 to 7 cumulative days preceding the fall with corresponding control periods selected 8–14 days before the fall. We used SAS version 9.2 for all analyses.

Additional Analyses

In order to address a limitation inherent to observational drug studies, confounding by indication, we performed two additional analyses. First, because trazodone may be used to treat acute behaviors other than depression, we repeated the analysis of non-SSRI antidepressants after excluding participants with a change in trazodone. Second, we repeated our analyses in two additional classes of medications not thought to be related to falls: acetaminophen and antibiotics.

RESULTS

Mean age of participants was 88 years (±SD 8 years), and 71% were females (Table 1). The majority of participants had moderate cognitive impairment (60.1%) and moderate functional impairment (53.8%). The most frequent comorbidities were hypertension (63.2%) and arthritis (31.2%). Forty-three percent of participants were taking more than 8 medications, and 57.1% were prescribed an antidepressant. Prescriptions for SSRI and other antidepressants were similar (32.5% vs 35.4%).

Table 1.
Characteristics of 1,181 Nursing Home Residents With a Fall in a Case-Crossover Study* to Examine the Acute Effects of a New Prescription or Increased Dose of an Antidepressant on the Risk of Falls (%, unless otherwise specified)

During the study, 70 participants experienced 86 changes of an antidepressant (non-SSRI changes = 59; SSRI changes = 27). Non-SSRI antidepressant changes included bupropion (n = 4), duloxetine (n = 1), mirtazapine (n = 21), trazodone (n = 29), and venlafaxine (n = 4). There were no changes of tricyclic antidepressants. SSRI antidepressant changes included citalopram (n = 18), escitalopram (n = 2), paroxetine (n = 2), and sertraline (n = 5). During the 7 days before the fall (hazard period), there were 22 new prescriptions for an antidepressant and 24 prescriptions for an increased dose compared with 21 new prescriptions and 19 increased doses during the 8–14 days before the fall (control period). Eight participants experienced an antidepressant change during both the hazard and control periods.

When all antidepressant changes were considered, the greatest odds of falls occurred within 2 days of a new prescription or increased dose of an antidepressant (odds ratio [OR]: 2.8, 95% confidence interval [CI] 1.1–7.2). The effect was attenuated and became nonsignificant when the analysis was extended to 7 days (OR: 1.4, 95% CI, 0.8–2.3). When we repeated the analysis among participants who received a new prescription or increased dose of an antidepressant separately, the results were similar (OR within 2 and 7 days following a new prescription: 2.0, 95% CI, 0.8–5.3 and 1.0, 95% CI, 0.5–1.8; OR within 2 and 7 days following an increased dose: 3.5, 95% CI, 0.7–16.8 and 1.5, 95% CI, 0.8–2.9, respectively).

The effect of a non-SSRI antidepressant change on the acute risk of falls was similar when medications were examined individually, although the numbers of these individual antidepressant drug changes were too small to provide meaningful estimates (results not shown). Thus, all non-SSRIs were combined in the analyses. For non-SSRI antidepressants, the maximum effect of falling occurred within 2 days of a new prescription or increased dose of the medication (OR: 4.7, 95% CI, 1.3–16.2; Figure 1). The OR of falling remained elevated 4 days after non-SSRI change, but it was no longer significant by 5 days (OR within 5 days: 1.9, 95% CI, 0.9–4.0). No increase in the risk of falls was found within 2 days of a new prescription or increased dose of an SSRI antidepressant (OR: 0.8, 95% CI, 0.2–3.4; Figure 2). Increasing the hazard window from 2 to 7 cumulative days had little effect on the risk of falls. (OR within 7 days: 0.7, 95% CI, 0.3–1.6).

Figure 1.
Odds of falling among nursing home residents with a new prescription or increased dose of a non-selective serotonin reuptake inhibitor antidepressant.
Figure 2.
Odds of falling among nursing home residents with a new prescription or increased dose of a selective serotonin reuptake inhibitor antidepressant.

When stratifying by baseline medication use, there was some suggestion that the effect of a non-SSRI antidepressant was greater among residents who used a large number of medications (OR within 3 days: 5.5, 95% CI, 1.2–24.8 in residents using a large number of medications and OR: 2.3, 95% CI, 0.6–9.0 in residents using fewer medications). Similarly, the effect of a non-SSRI antidepressant was greater among recent fallers compared with participants without a recent fall (OR within 3 days: 10.0, 95% CI, 1.3–78.1 among recent fallers and OR: 2.0, 95% CI, 0.6–6.6 among non-recent fallers). Interaction terms were not statistically significant (not shown).

Additional Results

After excluding participants with a change in trazodone, the maximum effect on falling occurred within 2 days of an increased dose or new prescription of a non-SSRI antidepressant (OR: 9.0, 95% CI, 1.1–71.0). The odds ratio of falling was no longer significant by Day 4 (OR: 2.3, 95% CI, 0.9–5.6).

No apparent association was found between a change in acetaminophen and risk of falls (OR within 2 days: 0.6, 95% CI, 0.1–2.5 and OR within 5 days: 1.2, 95% CI, 0.5–2.8). A new antibiotic prescription was associated with a nonstatistically significant elevated risk of falls that was relatively stable over time (OR within 2 days: 1.6, 95% CI, 0.9–2.8 and OR within 5 days: 1.4, 95% CI, 0.9–2.1).

DISCUSSION

We found that there was greater than a fivefold increased odds of falling within 2 days of a new prescription or increased dose of a non-SSRI antidepressant. With each day following the non-SSRI antidepressant change, the risk of falls diminished. We observed no association between the new prescription or increased dose of an SSRI antidepressant and the acute risk of falls.

Although many studies have examined chronic use of antidepressants as a risk factor for falls, few studies have considered the short-term effects of a change in an antidepressant prescription on the risk of falls. Two small case-crossover studies found a two to threefold increased odds of falling within 1–3 days of a new prescription or increased dose of a centrally acting medication (14,21). In both studies, the authors grouped multiple classes of medications together; thus, they were unable to determine the independent effects of an antidepressant change on risk of falls. Two larger studies examined the effects of a new antidepressant prescription on the risk of hip fracture (22,23). Liu and colleagues (22) found an increased risk of hip fracture within 30 days of a new prescription of a tricyclic or SSRI antidepressant. Using both a case-control and self-controlled case-series study design, Hubbard and colleagues (23) found a twofold increased risk of hip fracture within 2 weeks of a new prescription for a tricyclic or SSRI antidepressant. Although intriguing, these studies are limited as they relied on large administrative databases with less certainty about adherence and timing of the newly administered medicine, and they contained no information on non-injurious falls. Furthermore, they did not include trazodone or serotonin–norepinephrine reuptake inhibitors, which are commonly prescribed today.

Our findings suggest an association between initiation of a non-SSRI antidepressant prescription and the acute risk of falls. Most of the data linking non-SSRI antidepressants with falls is in relation to TCAs or trazodone (6). Little data exist to describe the association between newer agents, such as serotonin–norepinephrine reuptake inhibitors or mirtazapine, and the risk of falls (6,8,9). We speculate that the acute risk of falls that we observed with non-SSRI antidepressants may be mediated through acute cognitive or motor effects that have not been fully described to date. Future studies should examine the risk of falls among new users and chronic users of these drugs, as well as the mechanism associated with the increased risk of falls.

Our finding that SSRI changes are not associated with an acute increased risk of falls warrants further investigation. Previous studies have found that chronic SSRI use is associated with a similar risk of falls compared with use of traditional antidepressants (6,24). Hubbard and colleagues (23) reported an increased risk of hip fracture within 14 days of a new prescription for an SSRI antidepressant (OR case–control 6.3; IRR case-series 2.0). However, when restricting the analysis to adherent participants, who remained on drug treatment for at least 1 year, the authors found an increased risk of hip fracture only after 15–42 days of an SSRI prescription. It is possible that our failure to find an association between new prescriptions of an SSRI and falls is a reflection of the small number of SSRI changes we observed. However, it may be that these drugs increase the risk of falls only weeks to months after initiation, when therapeutic levels are achieved resulting in a decrease in depressive symptoms, increased physical activity, and subsequent opportunity for falls.

Our results suggest that the effect of a non-SSRI antidepressant change may be greater among residents using a large number of medications and those with a recent history of fall. These differences were not statistically significant likely due to small numbers. Nonetheless, it is plausible that these characteristics and others, including balance disorders and use of other psychotropic medications, may increase the vulnerability of nursing home residents to falls in the days following an antidepressant change. Identification of subgroups at high risk for falls is likely to increase the success of fall prevention strategies based on a medication change.

Our study has several limitations. First, falls may be underreported in a nursing home. It is doubtful that falls would be differentially reported based on medication exposure, and thus, underreporting is unlikely to have influenced our results. Second, medication changes were ascertained from computerized physician order entry without verification of administration. Third, although we recognize that the pharmacologic mechanisms of non-SSRI antidepressants vary, we had limited power to perform subgroup analyses to examine the acute effects of individual medications. We would encourage studies with large pharmacoepidemiology databases to examine both the acute and chronic effects of individual non-SSRI medications on the risk of falls. Fourth, we were not powered to consider the effect of dose, or timing of administration, on acute risk of falls. Of the 43 new prescriptions during the study period, 79% called for the lowest or below the lowest recommended starting dose of the medication. It is possible that the use of larger than recommended starting doses would have resulted in a stronger association with fall risk.

Fifth, we did not have information on the reason leading to the antidepressant change or concurrent medical illness. Although we selected a control period only 1 week before the hazard period, acute medical changes, such as delirium, may develop quickly in the nursing home setting such that it is difficult to discern whether the medication or the medical illness resulted in the increased risk of falls. On the other hand, depression is a chronic illness, and changes in antidepressants are not commonly made in the setting of acute medical illness. Because trazodone is often used to treat insomnia and agitation, we repeated the analysis of non-SSRI antidepressants after excluding participants with a change in trazodone, and the results were similar. Furthermore, when we repeated our analyses with acetaminophen and antibiotics, medications that are not traditionally related to falls but are often prescribed for acute medical conditions that can be associated with falls, we found no association or a nonstatistically significant association that was relatively stable over time. This suggests that our findings may not be explained entirely by the underlying medical illness but instead may relate to the medication itself. Regardless of our ability to separate the effects of the antidepressant change from the effects of the condition for which it was prescribed, our results are still clinically important as they identify a window of time during which nursing home residents are at high risk for falls, and interventions to reduce falls may be effective.

Finally, our study was performed in two nursing homes from a single organization. In comparison with other nursing homes, residents of Hebrew SeniorLife are more likely to be white (98% vs 88%) (25), private pay at the time of admission (39% compared with 28%) (25), and prescribed an antidepressant (57% vs 48%) (26).

In conclusion, our results identify the days following a new prescription or increased dose of a non-SSRI antidepressant as a window of time associated with a particularly high risk of falling among nursing home residents. Our results should be confirmed, but we recommend increased surveillance of nursing home residents in the days following a non-SSRI antidepressant change in an effort to reduce falls. Clinicians should avoid making changes around weekends or during times when unfamiliar staff is present. Future fall prevention programs should consider whether these measures are effective in reducing falls in the nursing home setting.

FUNDING

This research was funded by a Hartford Geriatrics Health Outcomes Research Scholars Award, a grant from the NIA (K23 AG033204), and the Men’s Associates of Hebrew SeniorLife.

Acknowledgments

This work was presented in part at the 2010 annual meeting of the American Geriatrics Society in Orlando, FL on May 15, 2010.

References

1. Rubenstein LZ, Josephson KR, Osterweil D. Falls and fall prevention in the nursing home. Clin Geriatr Med. 1996;12(4):881–902. [PubMed]
2. Norton R, Campbell AJ, Reid IR, et al. Residential status and risk of hip fracture. Age Ageing. 1999;28(2):135–139. [PubMed]
3. Rubenstein LZ, Josephson KR, Robbins AS. Falls in the nursing home. Ann Intern Med. 1994;121(6):442–451. [PubMed]
4. Injury Prevention & Control: Home and Recreational Safety. 2009. Centers for Disease Control and Prevention. http://www.cdc.gov/HomeandRecreationalSafety/Falls/nursing.html. Accessed February 25, 2010.
5. Agashivala N, Wu WK. Effects of potentially inappropriate psychoactive medications on falls in US nursing home residents: analysis of the 2004 National Nursing Home Survey database. Drugs Aging. 2009;26(10):853–860. [PubMed]
6. Hartikainen S, Lonnroos E, Louhivuori K. Medication as a risk factor for falls: critical systematic review. J Gerontol A Biol Sci Med Sci. 2007;62(10):1172–1181. [PubMed]
7. Woolcott JC, Richardson KJ, Wiens MO, et al. Meta-analysis of the impact of 9 medication classes on falls in elderly persons. Arch Intern Med. 2009;169(21):1952–1960. [PubMed]
8. Joo JH, Lenze EJ, Mulsant BH, et al. Risk factors for falls during treatment of late-life depression. J Clin Psychiatry. 2002;63(10):936–941. [PubMed]
9. Kallin K, Gustafson Y, Sandman PO, Karlsson S. Drugs and falls in older people in geriatric care settings. Aging Clin Exp Res. 2004;16(4):270–276. [PubMed]
10. Maclure M. The case-crossover design: a method for studying transient effects on the risk of acute events. Am J Epidemiol. 1991;133(2):144–153. [PubMed]
11. Mittleman MA, Maclure M, Glasser DB. Evaluation of acute risk for myocardial infarction in men treated with sildenafil citrate. Am J Cardiol. 2005;96(3):443–446. [PubMed]
12. Barbone F, McMahon AD, Davey PG, et al. Association of road-traffic accidents with benzodiazepine use. Lancet. 1998;352(9137):1331–1336. [PubMed]
13. Hunter DJ, York M, Chaisson CE, Woods R, Niu J, Zhang Y. Recent diuretic use and the risk of recurrent gout attacks: the online case-crossover gout study. J Rheumatol. 2006;33(7):1341–1345. [PubMed]
14. Sorock GS, Quigley PA, Rutledge MK, et al. Central nervous system medication changes and falls in nursing home residents. Geriatr Nurs. 2009;30(5):334–340. [PubMed]
15. Pharmacists ASoH-S. American Hospital Formulary Service. 2010. http://www.ahfsdruginformation.com/. Accessed March 8, 2010.
16. Morris JN, Hawes C, Fries BE, et al. Designing the national resident assessment instrument for nursing homes. Gerontologist. 1990;30(3):293–307. [PubMed]
17. Hartmaier SL, Sloane PD, Guess HA, Koch GG, Mitchell CM, Phillips CD. Validation of the Minimum Data Set Cognitive Performance Scale: agreement with the Mini-Mental State Examination. J Gerontol A Biol Sci Med Sci. 1995;50(2):M128–M133. [PubMed]
18. Katz S, Downs TD, Cash HR, Grotz RC. Progress in development of the index of ADL. Gerontologist. 1970;10(1):20–30. [PubMed]
19. Kiely DK, Kiel DP, Burrows AB, Lipsitz LA. Identifying nursing home residents at risk for falling. J Am Geriatr Soc. 1998;46(5):551–555. [PubMed]
20. Delaney JA, Suissa S. The case-crossover study design in pharmacoepidemiology. Stat Methods Med Res. 2009;18(1):53–65. [PubMed]
21. Neutel CI, Perry S, Maxwell C. Medication use and risk of falls. Pharmacoepidemiol Drug Saf. 2002;11(2):97–104. [PubMed]
22. Liu B, Anderson G, Mittmann N, To T, Axcell T, Shear N. Use of selective serotonin-reuptake inhibitors or tricyclic antidepressants and risk of hip fractures in elderly people. Lancet. 1998;351(9112):1303–1307. [PubMed]
23. Hubbard R, Farrington P, Smith C, Smeeth L, Tattersfield A. Exposure to tricyclic and selective serotonin reuptake inhibitor antidepressants and the risk of hip fracture. Am J Epidemiol. 2003;158(1):77–84. [PubMed]
24. Ensrud KE, Blackwell TL, Mangione CM, et al. Central nervous system-active medications and risk for falls in older women. J Am Geriatr Soc. 2002;50(10):1629–1637. [PubMed]
25. Gabrel CS. Characteristics of elderly nursing home current residents and discharges: data from the 1997 National Nursing Home Survey. Adv Data. 2000;25(312):1–15. [PubMed]
26. Hanlon JT, Handler SM, Castle NG. Antidepressant prescribing in US nursing homes between 1996 and 2006 and its relationship to staffing patterns and use of other psychotropic medications. J Am Med Dir Assoc. 2010;11(5):320–324. [PMC free article] [PubMed]

Articles from The Journals of Gerontology Series A: Biological Sciences and Medical Sciences are provided here courtesy of Oxford University Press
PubReader format: click here to try

Formats:

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...

Links

  • MedGen
    MedGen
    Related information in MedGen
  • PubMed
    PubMed
    PubMed citations for these articles

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...