Among 21,819 beneficiaries (Fig. 1), the median age was 58 (min 20, IQR 46–71, max 101; 39% were at least 65 years old), 50% were female, 84% were white, and 62% were dual eligible for Medicaid (Table 1).
Table 1
Population description (N = 21,819). ASM = antiseizure medication.
| | Median or N (interquartile range or %) |
Age | | 58 (46–71) |
Female sex | | 10,914 (50%) |
Race* | White | 17,809 (84%) |
| Black | 2,510 (12%) |
| Hispanic | 670 (3%) |
| Asian | 257 (1%) |
Dual eligible for Medicaid | | 13,717 (63%) |
Rural ZIP code | | 6,231 (29%) |
Reason for entitlement | Disability | 13,229 (61%) |
| Age | 8,581 (39%) |
| End-stage renal disease | 42 (< 1%) |
Region | South | 8,230 (39%) |
| Midwest | 5,644 (27%) |
| Northeast | 4,075 (19%) |
| West | 3,225 (15%) |
Any neurologist visit, 2015 | 15,133 (69%) |
Any primarily epilepsy visit, 2015 | 17,203 (79%) |
Unique medications (No.), 2015 | 8 (5–12) |
Unique ASMs (No.), 2015 | 1 (1–2) |
ASM proportion of days covered (%), 2014–2015 | 92% (81%-96%) |
Older generation ASM, 2015 | 12,517 (57%) |
Brand name ASM, 2015 | 5,430 (25%) |
Total part D out of pocket cost, 2015 | $66 ($0-$301) |
Epilepsy type** | Focal | 6,046 (28%) |
| Generalized | 4,270 (20%) |
| Both | 2,169 (10%) |
| Neither | 9,334 (43%) |
Refractory epilepsy, 2014–2015 | | 4,445 (20%) |
Epileptogenic | Ischemic stroke | 2,160 (10%) |
neurological conditions, | Traumatic brain injury | 563 (3%) |
2014–2015 | Intracranial hemorrhage | 516 (2%) |
| Tumor | 454 (2%) |
| Meningoencephalitis | 109 (< 1%) |
| Cardiac arrest | 49 (< 1%) |
Dementia | | 1,762 (8%) |
Charlson comorbidity index, | 0 | 12,938 (59%) |
2015 | 1–3 | 7,968 (37%) |
| 4–6 | 788 (4%) |
| 7+ | 125 (1%) |
Acute care visits, 2015*** | 0 | 17,005 (78%) |
| 1 | 3,267 (15%) |
| 2+ | 1547 (7%) |
Primary ASM prescriber, | Neurologist | 11,242 (59%) |
2015**** | Epileptologist | 430 (2%) |
| Primary care physician | 6,872 (36%) |
| Female | 4,315 (23%) |
| Years since med school | 27 (19–34) |
| Physician extender | 1,832 (9%) |
| D.O. | 1,515 (8%) |
| # visits this patient, 2015 | 2 (1–3) |
*Race: This is how Medicare classifies race, with Hispanic as a separate category without distinguishing non-Hispanic White versus Hispanic White versus non-Hispanic Black. |
**Epilepsy type: at least one International Classification of Diseases code for focal and/or generalized epilepsy. |
***Acute care visits: we excluded beneficiaries with epilepsy-related acute care visits, thus this variable refers to any non-epilepsy acute care visit. |
****Primary ASM prescriber: the single physician who prescribed the greatest number of antiseizure medication prescriptions and pill days in 2015. For the main definition, ‘epileptologist’ was defined as at least 25% of a provider’s Evaluation/Management codes being primarily for epilepsy, though we performed several sensitivity definitions of this (less restrictive: at least 10%; more restrictive: at least 25% plus at least 25 visits in the year). Physician extender was defined as Nurse Practitioner or Physician Assistant. |
The following competing risks occurred before the occurrence of a 180-day gap: 9,957 (46%) had at least one emergency room or inpatient visit listing epilepsy or seizures in any ICD position (median 362 days to the first visit, IQR 164–661), 358 (2%) died (median 531 days until death, IQR 346–758), and 179 (1%) had at least one month without continuous part D coverage (median 700 days to first loss of coverage, IQR 487–821). There were 2,536 (12%) beneficiaries with at least one emergency room or inpatient visit listing epilepsy or seizures in the primary ICD code position (median 464 days, IQR 209–752).
Overall, 5,191 (24%) had a 30-day ASM gap, 1,753 (8%) had a 90-day gap, 834 (4%) had a 180-day gap, and 381 (2%) had a 360-day gap.
Figure 2 displays cumulative incidence functions. On January 1, 2018, at which point all outcomes were measurable, cumulative incidences were 20%, 6%, 3%, and 2% for 30-, 90-, 180-, and 360-day gaps, respectively.
Among those 5,191 with a 30-day gap, 4,800 (92%) had a subsequent ASM fill during the 2016–2018 observation window. Percentages for other gap durations were 1,301/1,753 (74%) for 90-day gaps, 405/803 (50%) for 180-day gaps, and 102/381 (27%) for 360-day gaps.
In our sensitivity population definition, among the 19,010 (87%) with at least two ‘epilepsy’ (345.xx/G40.xx) ICD codes in addition to ASM fills during each of the baseline years (2014–2016), outcomes were nearly identical to the primary analysis: 24%, 8%, 4%, and 2%, respectively.
Our Cox regression demonstrated moderate concordance (0.72) and good calibration (Fig. 3). Table 2 displays adjusted predictors of 180-day gaps. Numerous variables predicted increased chance of a 180-day gap: number of unique medications in 2015 (hazard ratio [HR] 1.03 per medication, 95% confidence interval [CI] 1.02–1.05), meningoencephalitis (HR 2.44, 95% CI 1.10–5.40), and epileptologist prescribing physician (HR 2.38, 95% CI 1.42-4.00). Other variables predicted decreased 180-day gaps: Medicaid dual eligibility (HR 0.74, 95% CI 0.59–0.93), number of unique ASMs in 2015 (e.g., 2 versus 1: HR 0.37, 95% CI 0.30–0.45), greater baseline adherence (> 80% versus ≤ 80% of days in 2015 with ASM pill supply: HR 0.38, 95% CI 0.32–0.44), and older generation ASM (HR 0.81, 95% CI 0.68–0.97). Age was not a significant adjusted predictor (per decade: HR 0.98, 95% CI 0.88–1.06).
Table 2
Associations between each variable and a 180-day gap, N = 17,385. ASM = antiseizure medication.
| | Hazard ratio* | 95% confidence interval |
Age, per decade | | 0.97 | 0.88–1.06 |
Female sex | | 1.09 | 0.93–1.28 |
Race | White | Reference | Reference |
| Black | 1.14 | 0.90–1.46 |
| Hispanic | 1.16 | 0.74–1.83 |
| Asian | 1.62 | 0.96–2.78 |
Dual eligible for Medicaid | | 0.74 | 0.59–0.93 |
Rural ZIP code | | 0.99 | 0.83–1.19 |
Reason for entitlement | Age** | 1.16 | 0.87–1.54 |
Region | South | Reference | Reference |
| Midwest | 0.86 | 0.70–1.06 |
| Northeast | 0.86 | 0.67–1.09 |
| West | 1.02 | 0.81–1.30 |
Any neurologist visit, 2015 | | 0.90 | 0.72–1.13 |
Any primarily epilepsy visit, 2015 | | 0.88 | 0.72–1.07 |
Unique medications, 2015 | | 1.03 | 1.02–1.05 |
Unique ASMs (No.), 2015 | 1 | Reference | Reference |
| 2 | 0.37 | 0.30–0.45 |
| 3+ | 0.23 | 0.16–0.32 |
ASM proportion of days covered > 80%, 2014–2015 | 0.38 | 0.32–0.44 |
Older generation ASM, 2015 | 0.81 | 0.68–0.97 |
Total Part D out of pocket cost (per $100), 2015 | 1.00 | 0.99–1.01 |
Epilepsy type | Focal | Reference | Reference |
| Generalized | 0.99 | 0.77–1.26 |
| Both | 0.74 | 0.53–1.04 |
| Neither | 1.00 | 0.82–1.22 |
Refractory epilepsy | | 0.97 | 0.77–1.23 |
Epileptogenic | Ischemic stroke | 1.25 | 0.98–1.60 |
neurological conditions, | Traumatic brain injury | 0.76 | 0.44–1.33 |
2014–2015 | Intracranial hemorrhage | 1.26 | 0.82–1.94 |
| Tumor | 1.28 | 0.78–2.11 |
| Meningoencephalitis | 2.44 | 1.10–5.40 |
| Cardiac arrest | 0.59 | 0.17–2.03 |
Dementia | | 0.98 | 0.73–1.31 |
Charlson comorbidity index, | 0 | Reference | Reference |
2015 | 1–3 | 0.90 | 0.74–1.08 |
| 4–6 | 1.15 | 0.79–1.69 |
| 7+ | 1.31 | 0.65–2.67 |
Acute care visits for any | 0 | Reference | Reference |
condition, 2015 | 1 | 1.06 | 0.85–1.33 |
| 2+ | 1.07 | 0.80–1.45 |
Primary ASM prescriber | Neurologist | 1.06 | 0.86–1.30 |
| Epileptologist | 2.38 | 1.42-4.00 |
| Female | 1.03 | 0.85–1.25 |
| Decades since med school | 0.98 | 0.90–1.06 |
| Physician extender*** | 0.98 | 0.84–1.30 |
| D.O. | 1.11 | 0.82–1.50 |
| # visits this patient, 2015 | 1.03 | 1.00-1.06 |
*Hazard ratios are adjusted for all other variables contained in this table. Variables that appear in Table 1 but not Table 2 were omitted due to non-proportional hazards, unless mentioned below. Bolded hazard ratios are significant at p < 0.05. |
**End-stage renal disease was omitted due to unstable estimates with a small sample size and thus essentially reason for entitlement of age is essentially being compared with reason for entitlement of disability. |
***Hazard ratios for primary ASM prescriber were all computed in this model including only beneficiaries whose primary ASM prescriber was a physician, given data from the Physician Masterfile as covariates. Thus the main model did not include ‘physician extender’ as a covariate, but we did rerun the model omitting physician variables and including ‘physician extender’ as a variable (N = 19,729) with little meaningful change to other reported coefficients. |
Absolute effects were all small. For example, the frequency of 180-day gaps for selected comparisons were: 1) 5% versus 3% for age at least 65 years versus less than 65 (Fig. 4), 2) 5%, 2%, and 1% for beneficiaries on one, two, or at least three ASMs, 3) 3% versus 6% for those with PDC at least 80% versus less than 80%, and 4) 4% regardless of whether a beneficiary’s main prescriber was an epileptologist (though this was significant in adjusted analysis, Table 2).
We performed additional sensitivity analyses:
First, among the 7,672 (35%) with no refractory epilepsy code and proportion of days covered over 80% on monotherapy at baseline, the frequency of gaps of each duration was similar to the primary analysis (26%, 8%, 4%, and 2%, for 30-, 90-, 180-, and 360-day gaps, respectively; cumulative incidence functions essentially identical, not displayed). Among those with a refractory code 2.47% had a 180-day gap, and among those without a refractory code 3.99% had a 180-gap (unadjusted p < 0.01), though per Table 2 refractory epilepsy was not a significant adjusted predictor.
Second, having any dual Medicaid eligibility in 2016–2018 was associated with lower cumulative incidence of each gap (p all < 0.01). The probability of 30-, 90, 180-, and 360-day gaps for beneficiaries with versus without dual eligibility were: 19% versus 33%, 6% versus 11%, 3% versus 5%, and 1% versus 2%, respectively.
Third, we restricted to the top three ‘pure’ ASMs. Probabilities of 180-day gaps were 6% for levetiracetam, 5% for phenytoin, and 3% for phenobarbital.
Fourth, when modifying the definition of ‘epileptologist’ to be less stringent (at least 10% of visits primarily for epilepsy), ‘epileptologist’ in a regression otherwise identical to Table 2 was no longer significant (HR 0.88, 95% CI 0.60–1.28). When modifying the definition to be more stringent (at least 25% of visits primarily for epilepsy plus at least 25 visits primarily for epilepsy), ‘epileptologist’ remained significant (HR 2.23, 95% CI 1.26–3.99).
Fifth, when modifying our censorship procedure to consider the first acute visit listing epilepsy/seizures in the primary ICD code position (as opposed to any position), 7,036 (32%) had a 30-day ASM gap, 2,427 (11%) had a 90-day gap, 1,118 (5%) had a 180-day gap, and 531 (2%) had a 360-day gap before censorship.