Article Text

Download PDFPDF

Comorbidities in the first 2 years after arthroscopic hip surgery: substantial increases in mental health disorders, chronic pain, substance abuse and cardiometabolic conditions
  1. Daniel I Rhon1,2,
  2. Tina A Greenlee1,
  3. Bryant G Marchant3,
  4. Charles Dennis Sissel4,
  5. Chad E Cook5
  1. 1 Department of Rehabilitation Medicine, Center for the Intrepid, Brooke Army Medical Center, San Antonio, Texas, USA
  2. 2 Doctoral Program in Physical Therapy, Baylor University, Fort Sam Houston, Texas, USA
  3. 3 Department of Orthopaedics and Rehabilitation, Madigan Army Medical Center, Tacoma, Washington, USA
  4. 4 Program Analysis and Evaluation Division, US Army Medical Command, Fort Sam Houston, Texas, USA
  5. 5 Division of Physical Therapy, Department of Orthopedics, Duke University, Durham, North Carolina, USA
  1. Correspondence to Dr Daniel I Rhon, Department of Rehabilitation Medicine, Brooke Army Medical Center, San Antonio, TX 78234, USA; daniel_rhon{at}


Objectives We aimed to identify the rate of seven comorbidities (mental health disorders, chronic pain, substance abuse disorders, cardiovascular disorders, metabolic syndrome, systemic arthropathy and sleep disorders) that occurred within 2 years after hip arthroscopy.

Methods Data from individuals (ages 18–50 years) undergoing arthroscopic hip surgery between 2004 and 2013 were collected from the Military Health System (MHS) Data Repository (MDR). The MDR captures all healthcare encounters in all settings and locations for individuals within the MHS. Person-level data over 36 months were pulled and aggregated. Seven comorbidities related to poor outcomes from musculoskeletal disorders (mental health disorders, chronic pain, substance abuse disorders, cardiovascular disorders, metabolic syndrome, systemic arthropathy and sleep disorders) were examined 12 months prior and 24 months after surgery. Changes in frequencies were calculated as were differences in proportions between presurgery and postsurgery.

Results 1870 subjects were identified (mean age 32.24 years; 55.5% men) and analysed. There were statistically significant increases (p<0.001) proportionally for all comorbidities after surgery. Relative to baseline, cases of mental health disorders rose 84%, chronic pain diagnoses increased 166%, substance abuse disorders rose 57%, cardiovascular disorders rose by 71%, metabolic syndrome cases rose 85.9%, systemic arthropathy rose 132% and sleep disorders rose 111%.

Conclusions Major (potentially ‘hidden’) clinical comorbidities increased substantially after elective arthroscopic hip surgery when compared with preoperative status. These comorbidities appear to have been overlooked in major studies evaluating the benefits and risks of arthroscopic hip surgery.

Level of evidence Prognostic, level III.

  • arthroscopic surgery
  • hip
  • cohort study
  • health

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

What are the findings?

  • We captured the rate of change in seven key comorbid conditions after an orthopaedic surgical procedure of the lower extremity.

  • Every single one of the seven comorbidity categories assessed (sleep, cardiovascular, mental health, chronic pain, systemic arthropathy, substance abuse, and metabolic syndrome) rose substantially after surgery.

How might they impact clinical practice in the future?

  • These are typically not considered complications or consequences of surgery. They should be.


A common goal for undergoing orthopaedic surgery is to improve quality of life (QoL). Rates of medical complications such as infections, blood clots or fractures are often reported after surgery and tend to be inversely associated with QoL.1 Patient-related (eg, body mass index (BMI), sex) and surgery-related factors (eg, emergency vs routine, procedure type) associated with complications have been identified and linked to greater risks of morbidity and mortality.2–4 While most complications are easy to link to the procedure, much less consideration has been given to comorbidities that may develop over time, often related to undergoing surgery, and the lifestyle changes that occur afterwards. Although comorbidities can also impair patient health and QoL, they are not systematically tracked or prevented perioperatively.

It is not always clear if the comorbidity is caused by the index disease or vice versa. Just as the pre-existence of sleep disorders,5 mental health disorders,6 metabolic syndrome,7 substance use8–10 and systemic arthropathies11 can place patients at significant risk for physical complications and can impede recovery,12 developing one or more conditions as a consequence of surgery could also limit recovery.

Surgery to the lower extremity limits the patient’s ability to bear weight for weeks and can limit physical activity for months. Lack of exercise and mobility are associated with chronic diseases13 and excess time spent sitting or lying increases risk for metabolic syndrome and cardiovascular disease.14 15 Physical activity can help mitigate disease, minimise progression and improve QoL for individuals with chronic disease.16 17 We asked whether undergoing hip arthroscopy would make patients susceptible to developing additional morbidity.

Considering a potential relationship between surgery and subsequent incident morbidity, it is necessary to determine (a) to what extent comorbidities increase following surgery, (b) which conditions have higher proportional increases and (c) the relative risk of certain demographic factors for developing or exacerbating a condition. We aimed to identify the rate of medically diagnosed comorbidities that develop within 2 years following arthroscopic hip surgery; the secondary aim was to examine the change in comorbidities before to after surgery.

Materials and methods

Design and oversight

This was an observational cohort utilising claims data from within the US Military Health System. TRICARE is the payer of the healthcare system managed by the US Defense Health Agency (DHA). The REporting of studies Conducted using Observational Routinely collected health Data checklist,18 an extension of the STrengthening the Reporting of OBservational studies in Epidemiology ,19 was used to guide the reporting of this study and the study was approved by the Institutional Review Board at US Army Regional Health Command Central.

Patient involvement

Patients were not directly involved in the design of this study, primarily as it was retrospective in nature. The idea was generated from independent observations brought forth from multidisciplinary medical providers often co-managing these patients.

Data sourcing

The MHS Data Repository (MDR) was used to extract and identify the cohort and all healthcare utilisation variables for this study. The MDR includes exhaustive medical records documentation for individuals who are affiliated with all US armed services, to include active duty, reserve/national guard, retired service members and all family members and dependents. TRICARE services approximately 10 million unique persons in this setting worldwide. The MDR functions within a single-payer closed system and all healthcare interventions, both in military treatment facilities and out in the civilian network, around the world, are captured. Person-level data were abstracted and aggregated by a senior healthcare analyst working for the Army Medical Command with over 15 years of experience. Missing data were negligible and data validation procedures for this particular cohort have been addressed elsewhere.20

Participant selection

All individuals undergoing arthroscopic hip surgery over a 10-year period (July 2004 through July 2013) were identified with the presence of the following Current Procedural Terminology codes: 29862, 29914, 29915 and 29 916. Healthcare utilisation data were extracted for the 12-month period before surgery, and any individuals with a diagnosis of malignant neoplasm, osteoarthritis, infection, fracture or avascular necrosis of the hip in this time period were excluded. Individuals <18 and >50, anyone with history of any hip surgery in the previous 12 months and anyone without at least 2 years of eligibility for healthcare services after surgery were also excluded. Two-year follow-up is considered the standard minimum follow-up time for orthopaedic surgery studies, and 1 year prior would seem to identify the most relevant comorbidities the closest to the date of surgery. Because typical enlistment periods are 4 years, a broader surveillance window would have decreased the sample size substantially. Exclusion of prior medical cases and age was necessary to create a homogeneous cohort of patients most likely undergoing arthroscopy for femoroacetabular impingement syndrome (FAIS), as that condition does not have its own International Classification of Diseases (ICD) code. For this final cohort, all healthcare utilisation 12 months before and 24 months after surgery was extracted (figure 1) from a period of 1 July 2003 through 30 July 2015.

Figure 1

Flow diagram of cohort selection. FAIS, femoroacetabular impingement syndrome; MDR, Military Health System Data Repository.

Individuals were categorised according to socioeconomic status (SES) and age, which can influence comorbidity rates.21 22 In the military, officers have a minimum 4-year Bachelor’s degree. The base pay for officers is also higher than for enlisted service members. Individuals were classified by household, where for example the spouse of an enlisted service member was also classified as enlisted.

Patient involvement

Patients were not involved in the design of the study.


Comorbidities were clustered into seven categories: metabolic syndrome, systemic arthropathy, disorders of mental health, chronic pain, substance abuse, cardiovascular system and sleep. The comorbid diagnoses were identified using reported medical ICD codes, ninth edition. Because many of these comorbidities are likely to accumulate over time as a potential consequence of surgery, we chose a 2-year period of surveillance after surgery. To determine presence of a condition prior to surgery, diagnosis codes appearing during the preceding 12 months were examined. We reported counts for both total cases and new cases (not present in the 12 months prior). Specific ICD-9 codes utilised for each comorbidity category and the rationale for inclusion have been published elsewhere.20

Statistical analyses

Descriptive statistics were reported and the prevalence and incidence of comorbidities after surgery were explored using raw changes in frequency before to after surgery. Differences in proportions between presurgery and postsurgery were also calculated with the McNemar Chi Square test with Yates correction and binomial distribution; significance level was set at p<0.05. Relative risk (RR) and 95% CIs were provided. Findings were grouped by age and SES, because comorbidities can be influenced by age21 and SES.22 All analyses were run using SPSS V.24.0.

Source of funding

This study was funded by the US DHA. They had no role in the study design, methodology, results or dissemination plan.


There were 1870 participants in the final cohort (figure 1), mean age 32.2 years, primarily men (n=1037, 55.5%), on active duty (n=1263, 67.5%) and lower SES (enlisted n=1415, 75.6%). About half of all surgeries occurred in a military facility (n=967, 51.7%) compared with a civilian network hospital (table 1). All subjects had complete follow-up data to 2 years.

Table 1

Demographic and clinical care variables for patients with and without postsurgical comorbidity

There were statistically significant increases (p<0.001) proportionally for every measured comorbidity within 2 years after surgery (table 2). The total number of individuals with a chronic pain diagnoses increased more than any other comorbidity (166.1%) followed by PTSD (149.3%) and the presence of any sleep disorder (111.3%). The comorbidity with the greatest increase of new cases was that of mental health disorders (26% of the entire cohort). Of the PTSD cases alone, 71.3% of cases present after surgery were new. 80.8% of chronic pain cases present after surgery were new cases. The percent of the entire cohort with any comorbidity present increased from 52% to 72%, with an incident increase of 53.5% (figure 2).

Table 2

Comorbidity measures before and after hip surgery

Figure 2

Rate of comorbidity change before to after surgery: new and persistent cases.

Patient age

Age was associated with RR for receiving a diagnosis for certain comorbid conditions (figure 3). Younger age (18–30) was protective for metabolic syndrome, cardiovascular disease, sleep disorders in general and sleep apnoea compared with the two older age groups. The oldest group (ages 41–50) had the highest RR for developing metabolic syndrome, cardiovascular disease and sleep disorders in general, compared with the two younger age groups. The RR of developing sleep apnoea was significantly elevated for the two highest age groups compared with the youngest age group.

Figure 3a

Relative risk of developing a comorbidity based on age and socioeconomic status.

Socioeconomic status

SES was also associated with RR for a number of comorbidities (figure 3). Individuals in the enlisted category were at significantly higher risk for a comorbid substance abuse disorder compared with the other SES categories. The officer category was protective for both substance abuse and mental health disorders compared with the other SES categories. Junior SES (officer or enlisted) was protective for sleep apnoea compared with the other SES categories. Senior enlisted had the highest RR for developing metabolic syndrome and cardiovascular disease compared with the other groups, but senior officers did not have this same risk.


Incidence of seven major comorbid conditions (mental health disorders, chronic pain, substance abuse disorders, cardiovascular disorders, metabolic syndrome, systemic arthropathy and sleep disorders) increased significantly after surgery. There was a 20% increase in the prevalence of any comorbidity after surgery. Importantly, the MDR is a single payer source that does not suffer from the monitoring effect seen in many prospective trials. In other words, the increase in comorbidities after surgery is not related to the way the data were collected. Further, 53.5% of individuals (n=480) without any of the specified comorbidities before surgery had at least one new comorbidity afterwards. Certain demographic factors (ie, age or rank) were linked to greater risk of developing certain comorbidities. Lastly, having surgery did not relieve comorbid conditions. The fact that the majority of patients (89.1%) with presurgical comorbidities had comorbidities 2 years postsurgery provides an important therapeutic target; orthopaedic surgeons may consider working more closely with internal medicine specialists to ensure appropriate management of these comorbid conditions in the years following the index orthopaedic operation.

Age and SES have been associated with a higher rate of comorbidities.23 With the passage of time, there is a greater opportunity for comorbidities to develop. In the MHS, the proportion of younger and enlisted individuals is much higher than that of older individuals and officers. Our findings were mixed depending on the type of comorbidity. Older individuals are much more likely to develop disorders of chronic disease (metabolic syndromes, cardiovascular disease and sleep disorders; primarily apnoea). A mental health or substance abuse disorder was much more likely to be present for someone in an enlisted household. It is possible that more serious cases of these disorders would have resulted in the end of a military career, selectively removing cases before they reached a higher rank. Disorders such as sleep apnoea and metabolic syndromes, such as diabetes, hypertension, and so on, are found at higher rates in older populations, so these findings appear intuitive. It is possible that these specific conditions are exacerbated due to changes in lifestyle related to surgery, or they could just be expected to occur with time, and why an age-matched non-surgical cohort would be important in order to make more definitive conclusions about the relationship of surgery and comorbidities. Chronic pain and substance abuse are relevant problem in military populations,24 25 and their relationship with surgical procedures should be further examined.

An increase in mental health disorders after hip surgery could have several explanations. First, this is a relatively extensive and traumatic procedure. Post-traumatic stress symptoms have been shown to develop after orthopaedic spine surgery, and are associated with greater disability.26 There is likely a loss of independence and self-reliance (ambulation is limited due to weight bearing restrictions, unable to drive and so on), In what is likely a very active population, it is probable that physical activity and exercise levels decrease immediately after this surgery. Exercise can counteract anxiety, stress and depression.27 The development of psychosocial risk factors and depression are associated with persistent pain after orthopaedic surgery.28 Patients in the MHS with mental health disorders had worse outcomes and were less likely to return to duty after hip arthroscopy for FAIS.29 However, less work has been done to assess the rate of mental health disorders developing after surgery. In one study, 56% of patients undergoing elective shoulder surgery had preoperative symptoms of psychological disorders which persisted after surgery, and 10% developed new symptoms of distress, somatisation, depression and anxiety 12 months after surgery.30 In our cohort, 79% had a preoperative mental health diagnoses still present within 2 years after surgery (15.7% of entire sample), and 26.1% of the entire sample developed a new mental health diagnoses (6.6% were PTSD). Youngest age or junior SES groups were linked to greater RR of substance use and mental health disorder development. This corroborates findings from other military cohorts, showing a higher risk of alcohol, cigarette and illicit drug use in lower pay grades.31

Poor sleep is a current problem in military populations,32 and has a bidirectional relationship with pain.33 Increased pain, common after a traumatic surgery, affects quality sleep and poor sleep lead to decreased thresholds for pain. After hip surgery, the patient has to be cautious of sleep positions. Breathing-related sleep disorders have been independently associated with pain syndromes, anxiety, PTSD, metabolic syndrome, diabetes, hypertension, stress and altered immune function.32 34 Patients undergoing shoulder surgery demonstrated sleep disturbance beyond 2 years, despite improvements in function and pain.35 Sleep disturbance is often reported after major operations.36 Even acute (7.33±2.53 d) hospitalisation for elective cardiac or orthopaedic surgery is linked to a twofold increase in insomnia cases 3 months postdischarge.37 Better quality sleep promotes daytime physical activity,38 and poor sleep could adversely influence exercise performance, limiting the physiological and cognitive benefits of exercise.39 Meeting sleep guidelines is associated with decreased risk of all-cause morbidity and better health-related QoL.40 In our cohort, clinical sleep disturbance appeared after hip surgery in 22.9% of the patients without prior history; older individuals were at higher risk.

Perceived pain and stress surrounding injury, surgery and recovery make patients vulnerable to sedentary activity, drug exposure and suppressed immunity.41 Clustering of comorbidities is also expected when resilience or vulnerability is altered42 and many of the conditions examined in our study have been reported to coexist.43 44 Lack of physical activity presents itself as a viable mediating factor between hip surgery and medical comorbidity. A bidirectional relationship between physical activity and pain has been reported, where higher levels of pain predicted lower levels of physical activity, but higher levels of physical activity predicted lower levels of pain.45 Incidence of cardiovascular disease, type 2 diabetes and cancer have all been associated with greater sedentary times.46 Sedentary behaviour has been linked independently to specific and all-cause mortality.47 Perioperative impacts on physical activity may stem from temporary medical restrictions, fear of reinjury or movement,48 49 and pain. Recent systematic reviews have reported no increase in physical activity 6 months following surgery and no or slight increases in the first year.50 51 In fact, even though joint replacement patients reported increases in physical activity, parallel accelerometry data was contradictory52; patients spent 83% of the day in sedentary behaviours 6 months postsurgery. Participating in physical activity, on the other hand, has been associated with better health and QoL53 54 and can ameliorate prescription drug use.55 Exercise therapy in the perioperative period has been recommended to help improve outcomes for patients with comorbidities.56

Generally, elective orthopaedic surgeries are same day procedures. Patients typically expect good prognosis (eg, return to work and physical activity, improved QoL and reduced pain).57 58 However, these findings indicate that some patients may be at risk of developing other medical conditions as a potential consequence of surgical interventions. These increase the complexity of future clinical care.42 It is possible that some of these conditions are short-lived and self-resolving after normal physical function, diet and sleep are restored; however, it is likely that some individuals take longer to return to this level of restored function and health. At the same time, there may be some individuals who see a reduction in comorbidities after surgery. For example, someone can now sleep when previously unable to sleep because of pain. There was a number of individuals with a comorbidity prior to surgery that did not have one after surgery (figure 2). Regardless, risk for comorbidity should be identified and ameliorated during the perioperative care period. Examples include educating patients about risk of comorbidities instead of only focusing on potential complications, as well as screening for sleep and mental health disorders. These results present opportunities for validation of these findings in future prospective trials, to include other healthcare systems and other patient populations. In addition, future investigation is needed to better understand how outcomes might improve when comorbidities are screened, considered by the medical team, and managed as part of the perioperative care process and surgical candidate selection.


This is the first study to identify rates of comorbidity before and after elective hip surgery. As we used observational data, these results do not imply causality. Further, the results of any observational research rely heavily on the interpretation of the researchers and can be influenced by confounders beyond statistical adjustment.

Subgrouping by severity of comorbidity could provide additional value, but not possible with these data. Patients were categorised as having the condition if at least one code was present in their medical records. At the same time, the threshold for comorbidity identification is based on patients seeking care and receiving a medical diagnosis. Early stages of comorbidities or those that appeared subclinical to an individual may not have resulted in care seeking and subsequent diagnosis. It was also not possible to determine if comorbid cases present prior to surgery had resolved after surgery. Their absence after surgery only indicates that patients did not seek care for them. It is unknown how many of these comorbidities would have developed regardless of surgery. The diagnosis codes reflect the level of accuracy for how they were entered into the electronic medical record and could vary by provider and setting. ICD-9 codes lack diagnostic specificity. For example, the same code is used for recurrent hypersomnia, menstrual-related hypersomnia and Kleine-Levin syndrome. These results may also not be the same in other populations (eg, non-military) or in other body regions, particularly non-weight-bearing regions such as the upper extremity. Finally, a non-surgical control group from the same setting is necessary to better isolate the true influence of surgery on these downstream comorbid conditions. With this particular condition, the primary challenge in establishing a control group is that the time from symptom onset to diagnosis can run an average of 32 months,59 and therefore establishing a common starting point for all patients would be very difficult.59


Our study suggests that comorbidities increase in the 2-year span following hip surgery. Hip arthroscopy is not a benign procedure and our study flags the need for treating clinicians to consider more comprehensive patient care approaches in the perioperative period and out to at least 2 years.


The authors would like to thank Drs Katie Dry, Rachel Mayhew and Laurel Proulx for their assistance with the data preparation for this analysis. They would also like to thank the Geneva Foundation for supporting research in military medicine.



  • Contributors DIR: received funding for the study; CDS: developed the extraction syntax and was responsible for data preparation for analyses; DIR, CEC and BGM: were involved in study question and design; DIR, CEC and TAG: were involved with analysing the data. All authors were involved and approved primary manuscript and revisions.

  • Funding This research was supported by an internal grant from the US Defense Health Agency (W911Qy-15-1-0016).

  • Disclaimer The view(s) expressed herein are those of the author(s) and do not reflect the official policy or position of Madigan Army Medical Center, Brooke Army Medical Center, the U.S. Army Medical Department, the U.S. Army Office of the Surgeon General, the Department of the Army, the Department of the AIr Force, or the Department of Defense, or the U.S. Government.

  • Competing interests None declared.

  • Patient consent Not required.

  • Ethics approval US Army Regional Health Command - Central IRB.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Data sharing statement Data belong to the US Defense Health Agency and can be shared if proper Data Sharing Agreement Applications are filed and approved through the Privacy Board Data Sharing Office at DHA.